Ever considered how effortlessly your brain calculates the dauntingly complex differential equations involved in deciding whether it is safe to cross the road when there is a car coming towards you at 30 miles per hour and a bicycle coming at 10 miles an hour in the other direction? You factor in a margin of safety and usually you get across without getting hit by the car or causing the cyclist to swerve violently, and the whole incident is a lesson in the efficiency of biological computing; Dr LeonardAdleman of the University of Southern California at Los Angeles is causing some excitement with a successful attempt to harness that extraordinary computing power by using a drop of dioxyribonucleic acid, DNA, to solve a seven-city iteration of the travelling salesman problem – the most efficient route to take to visit every city on the itinerary: Dr Adleman, who described his experiment in the November 11 issue of Science, says that molecular computers are breathtakingly fast and efficient, and have unheard-of storage capacities, running at more than 1 Tera-operations per second while using one thousand millionth as much power and storing 1Tb in the space required to store one byte in an electronic computer; Adleman used DNA segments to help direct the search process, finding the best solution from 1,000,000m possibilities; so will we see biological computers? Almost certainly we will, although the first ones are likely to be very problem-specific.