The art of aerodynamics becomes a tad fuzzy after that. Most F1 cars were designed by what Tony Southgate calls 'eye experience', plus calculation sometimes., The Lotus 16, the 'Mini Vanwall', was roughed out by Colin Chapman on the back of an envelope and run by Charlie Williams of Williams & Pritchard, who then made Lotus aluminium bodies. Costin had nothing to do with it. Which is why the drivers baked if the car lasted any length of time, which was not often.

The only F1 cars I know for certain saw the inside of a serious wind tunnel before the late 1970s and the arrival of ground effect were early Brabhams. Jack Brabham had all sorts of deals going, and one was with the Rootes Group, maker of Hillman, Sunbeam, Singer and Humber cars. Rootes had access to the full scale wind tunnel at MIRA (Motor Industry Research Association) and while the company was a stranger to aerodynamics, every so often Jack was allowed to put a car in the tunnel on the back of the Rootes connection. Thus it was that Ron Tauranac was able to confirm his hunch: outboard suspension added almost nothing to a car's drag, but it saved much time in setting up a when a customer arrived at a new circuit and had limited practice.

Brabham racing cars also owed something to a tip that Jack received from Malcolm Sayer who suggested the overhanging lip on the nosecone, Jack had bought a Lotus 24 to use until the Brabham BT2 was ready and the Lotus 24 was a very unstable car. The Lotus 25 worked because it had a slightly different nosecone. The Lotus 24 and 25 had similar suspension layouts, the main difference was the 24 had a spaceframe and the 25 had a primitive monocoque section. The 25 proved to be superior so the chassis has always got the credit. I submit that the margin of superiority of the Lotus 25 over the Lotus 24 was not the bit to which the suspension was anchored, but the shape of the nosecone.

In 1962, with 1.5 litre engines, the main aim was low drag. Part of that was the smallest possible radiator aperture. It did not occur to anyone that a few degrees difference in the attitude of the aperture to the air could make a great difference in handling.

Wind tunnels came into regular use after Peter Wright, working at Lotus, instigated the ground effect revolution. Ground effect was known in the aero industry and had been used on the Fleischer Storch a high winged monoplane with a stall speed of around 37 mph. With a high head wind, this plane could fly backwards. I have seen all kinds of aircraft going fast, but nothing is as impressive as a Storch hanging in the air.

There were two wind tunnels available in the UK around 1980, one was at Southampton University and the other was at Imperial College, London. Our jaws dropped when we heard that Frank Williams had booked Imperial College so nobody else could get in. For me, this was the start of modern F1, when a guy booked a wind tunnel, at £1,000 per day, so nobody else could use it. At that point I knew that Frank was going to be one of the greats.

When people added wings in F1 (well after Jim Hall and his Chaparral sports racers), they were mainly designed by guesswork. You or I could design a wing which would have improved a 1960s F1 car buy a model aeroplane kit and turn the wing upside down.

Until the mid 1980s wind tunnel testing, when it happened at all, operated on simple lines: car up, car down. nose up, nose down. It was John Barnard and Alan Jenkins, working at McLaren, who first devised proper wind tunnel programmes. For a start they understood that the radiator inlets had to be changed depending on the ambient temperature if the engine was to be cooled to the optimum. If the radiator inlets are changed, that has an effect on air flowing behind and they also understood that the two main things they could do was to keep the tyre happy and direct air under the rear wing in the most efficient way.

In the mid 1980s March built a wind tunnel which was cracked up to be the best in the world. It had everything including a rolling road which was not a given at the time. Because it was intended to be a profit centre, March Grand Prix continued to use the tunnel at Southampton University and came up with the March 881, a ground breaking design by Adrian Newey and Nick Wirth. Lotus hired the March tunnel and hit trouble. The problem was the insulation which was fourteen thousand times too efficient and the result was a pressure cooker.

Great ads never made: We had Honda engines and Ayrton Senna, then we used the March Wind Tunnel.

In 1989 Alan Jenkins became Chief Designer for the Onyx GP project and he invested in a state of art 40% wind tunnel model with all the data acquisition stuff, which is the real expenditure. It cost £350,000, then an astonishing amount, and two men were employed full time to keep the model equipped with bits. Stephan Johansson stood on the podium in Portugal and yet Onyx had to run in pre-qualifying in the first half of the season.

After that, I lost touch with aerodynamics. I had grasped the basics, as one does when staying up late with Frank Costin as the Scotch goes down. From the early 1990s it was all PhD stuff and the point of a PhD is that you end up knowing more and more about less and less. This is not meant to be pejorative, I have a PhD, but there are teams of very bright people working on the human genome who will never know fame outside of the research community. William Harvey, who discovered the circulation of blood in the human body, became famous.

Circulation of the blood is something we all know about and take for granted, it even seems odd that it took so long to discover You and I can understand Harvey's discovery, but what of the research now ongoing into gene research? Think of Frank Costin as Harvey and the teams of very gifted engineers working in wind tunnels as the very bright people who are looking at material we cannot see with the naked eye. Harvey sent medicine in a new direction and saved the lives of many a Pitpass reader, it is the researcher with a pharmaceutical company who is going to prolong our lives.