The two project students I've been supervising over the last few months are at opposite ends of the scientific aptitude spectrum. One is a born scientist. You know, methodical, switched on, careful with expensive equipment. The other is not quite there yet. I'm being charitable. I've had to learn fast how to find time in my own experiment-filled day to supervise someone who is a genuine help and regularly turns out useful results, along with someone who, although it isn?t their fault, is an unproductive drain on my time and energy.

Your boss is probably an old hand at dealing with project students. I know mine is. His approach is to pour attention on the "natural scientist" type of student and leave the other to flounder. But what if some of the so-called no-hopers just need a little encouragement? For instance, what if your apparently world-class slacker is paralysed deep down by a secret fear of failure. Or maybe the student who insists on ploughing a lonely, unproductive furrow is just embarrassed at the thought of asking you yet another 'stupid' question. Whoever I?m working with I make an effort to get to know them a bit. A little genuine interest goes a long way and if offered early enough might iron out issues before they become problems. I've witnessed a project student who looked unlikely to produce anything coming good in the end largely thanks to one kind person who noticed their predicament and, crucially, was prepared to give up quite a bit of their time to help.

If a potential winner selects your project then you're in luck. Devote your time to them and spend some of your grant money while they're still available. But despite your best efforts there will always be a few individuals whose general incompetence (breaking things), unreliability (not turning up), and/or low productivity (not comprehending the concept of a full day's work) persist. At the end of the day it's their degree so you'll have to resign yourself to simply moaning to your boss and putting up with them until they move on. In truth, most students will be between these two extremes: some natural aptitude but a great need to acquire science-related skills.

It goes without saying that the few hours of effort a month it takes to keep your materials and methods organised (and bang up-to-date) offers a massive time saving compared to tracking down your notes from scratch for each new student. In addition, writing down brief instructions specific to each project is likely to prevent at least a few minor queries later on. It will give the student some of the useful little knacks that you wouldn?t dream to include in your official M&Ms because an adept scientist would already know them. Stuff such as exactly how to hold the instrument at the right angle, that sort of thing.

Also, regularly drawing up a list of outstanding tasks or experiments, say each week, will help your student to remember what they are supposed to be doing. Try to keep the information content down to achievable short-term targets. I know that anything more would have overwhelmed me. Lists written up with my boss certainly helped me during my loneliest moments on the rock-face of my steep PhD learning curve. For want of a better word, a list can also form a sort of 'contract', laying down what you and your student agreed they would do. This will also help you to keep track. With my head so full of my own research I often completely forget the experiments I suggested to my students a week ago.

Foreign students for whom English is not easy offer an additional challenge to the fresh-faced supervisor. I recently had to explain exactly how a new piece of software worked to a student with an apparently fair ability to understand English but whose pronunciation was painfully difficult to follow. Even if I really concentrated on what they were saying I could only be sure of two words out of every three. Writing it all down was the only way we could get anywhere.

Drum in the need for replication. For some reason the need to overcome the vagaries of Nature by doing it again and again and again is rarely intuitive in students, myself included I might add. Also stress the need for accurate recording of the outcomes, even for ideas that fell flat on their face, as well as the importance of labelling which sample is which. Poor or, even worse, no labelling has an unremarkable ability to cause misinterpretation of results and unnecessary repeating of perfectly good experiments. Its incidence amongst novice scientists is endemic. I admit that I often used to think I'd simply remember what tubes A and B were!

The nature of research often makes it impossible to have set times when you can deal with queries. Dealing with unexpected interruptions with good humour is essential for a happy lab life. Take it as a compliment that your advice and expertise are in need. It always makes me realise just how far I've come. Try to keep your interventions brief and give your student just enough to get right back on the scientific ?horse? once they?ve fallen off. But despite all your detailed notes and many words of encouragement, the hands-on nature of science dictates that you'll often have to get, well, hands-on. Just the other day I stepped in when several attempts at verbal guidance had produced nothing. After 20 minutes we had solved the problem and, to boot, noticed a new phenomenon well worth pursuing. When it comes off, this sort of kick-starting adds a real buzz to project supervision. One of your most important jobs as a postdoc is to be a problem-solver, a trouble-shooter.

Above all, remember that it?s their project too, even though in some cases you might feel that you are more motivated and interested in the results than your student! If the project?s only value is that your student learns how not to do science, then your time and effort may have been worth it. If they are anything like me, it?ll save them wasting the start of their PhD by making all the simple mistakes.