But, as in so many aspects of education, the raw numbers do not give the full story. In 1988, the Stone Committee on the future of university chemistry was puzzled to find that the output of first-degree graduates fell year by year from 1969 to 1977 by about 30 per cent in total, and then recovered from 1978 to 1984, only to drop again. What was it we were doing right in the good years, they wondered, that we were not in the bad.
In fact, both the Stone Committee and my friend had overlooked the importance of fluctuations in the birthrate. One of the remarkable things about the numbers of A-level and degree students in physics and chemistry is how consistent they have been relative to the age cohort.
From the first record in 1962, the proportion of 18-year-olds specialising in the sciences has steadfastly remained at about 5 per cent. Interestingly, a similar proportion holds in the United States. In this country, science A-level entries have been affected neither by the switch to comprehensive schools nor the introduction of GCSEs, both of which have boosted other A-levels. None of the changes to course content, teaching methods or exam format seems to have made much difference. Physics and chemistry simply have not grown with other subjects. Should we worry?
From the point of view of finding the science professionals of the future, it may matter less than is commonly supposed. There seems to be a hard core of young people attracted to the sciences come what may. International comparisons suggest that they may be enough to meet Britain's needs.
Britain, in fact, has one of the highest outputs of science graduates in the world relative to its population, up there with Korea and Japan. But when it comes to employing them as science professionals, because of the comparatively low investment in research and development, it falls behind the United States, Germany, France and many other countries.
It is true that a number of university physics and chemistry departments are finding it difficult to fill their places and several are threatened with closure. But physics, and to a lesser extent chemistry, is the only A-level subject to have shown the demographic dip from 1983 forecast for all subjects.
The low-water mark for 18-year-olds was in 1995, and by the end of the century they will have increased by 12 per cent. If the demographic connection holds, we can therefore look to more physics and chemistry applicants, and we must keep enough slack in the university system to accommodate them.
It is not all good news, however. The relationship between the double-award science GCSE and physics and chemistry A-levels must be watched because the expected upturn in 1996 with the number of 18-year-olds did not occur. And there are two major problems which must be tackled. One is the difficulty of recruiting sufficient science teachers and the other, perhaps not unconnected, is the low level of scientific understanding of the person in the street.
The problem of attracting science teachers is so long-standing and deep-rooted that it almost certainly requires a structural change. Many reasons for recruitment difficulties have been discussed but none has led to a practicable solution. One that has not received much attention, however, is losing touch with the subject.
When I was a research scientist, what turned me on was knowing that I was at the cutting edge. The subject was alive for me and I was where it was at. When I moved across to education, I was shocked by how quickly I became out of date. While, at first, it was a pleasure to pass on what I knew, soon I was recycling old and second-hand material.
If my experience is anything to go by, I would have thought an experiment worth trying would be for several university physics and chemistry departments (which, after all, have spare capacity) to take over the training of science teachers.
Instead of students transferring to education departments for their PGCE year in what are often poorly-equipped laboratories staffed by one or two ex-teachers, it could be incorporated into the fourth year of new MPhys or MChem courses carrying qualified teacher status. Given the increased role of schools in teacher training, it should not be too difficult to organise.
The great advantage of this approach to those thinking of becoming teachers is that, especially if it were backed with the prospect of regular sabbaticals in university science departments, they could remain in their subjects. They would not have to decide to become teachers any earlier than now, but if they did so they could be assured that they would not become semi-detached, or out-of-date, or were committing themselves to a less-regarded option.
With a closer association of university and school science, cutting out the middleman as it were, we should not only be able to attract more but better teachers. This should lead in time to improved public understanding of science. It could even lift A-level physics and chemistry entries above the demographic trend.
Alan Smithers is professor of policy research and director of the centre for education and employment research at Brunel University