Spotlight on science: Tessa Blackstone defends her view that the proportion of students studying science and engineering should fall. Readers respond. Peter Swinnerton-Dyer suggests a different solution

Some 20 years ago, Brian Pippard, then Cavendish professor of physics at the University of Cambridge, suggested replacing the three-year first degree by a two-plus-two-year structure; at the end of the second year the less good students would leave and the better ones would go on to the remaining two years of a four-year course.

Pippard's experience had shown him that in physics the scheme he proposed would be an improvement on the one it was to replace. On the one hand, the amount of material that merited inclusion in a first degree course for good students was steadily increasing; on the other hand, a significant number of students reached the limit of their abilities by the end of their second year. Pippard's proposal received some support from physics and neighbouring disciplines; but it was vigorously opposed by other disciplines, which saw no advantage in the change.

Nowadays it is recognised that different disciplines have different characteristics and may benefit from different structures, in particular the structure of first degrees. This and other developments make it appropriate to take a further look at Pippard's proposal, though only in mathematics, physics and engineering. What these three subjects have in common is that they are all cumulative: in particular, the third year's work depends on the students having fully understood the material in the first two years.

ÖйúAƬ

Because of this, there are a significant number of students in these disciplines who benefit hardly at all from the third year of their course. Many of them know by the end of their second year that this will be so; but they feel that they have to struggle on, because otherwise their careers will be blighted by the stigma of having "dropped out". What they want, and what is not available to them, is an honourable exit.

Recent changes in student funding make their dilemma more acute, because staying on for a third year will considerably increase the debt with which they are burdened when they take jobs. Of course, there is no reason why students who leave after two years should not later return and take the third-year courses - or even the masters courses if they are qualified for them - either on a full-time or on a part-time basis. Realistically, however, this would only be practicable for those whose careers have kept them in touch with their discipline.

ÖйúAƬ

Pippard's proposal in its original form is no longer viable. The additional debt that each year of university study involves means that one cannot simply offer students in these disciplines the choice between two-year courses and four-year ones. But it is still desirable to provide an honourable exit after two years, and the growth in the number of four-year courses in these disciplines implies that there is a genuine demand for four-year courses.

I therefore suggest that in these three disciplines the degree structure should be changed, so that at the end of their second year students have these three choices:

* to leave the university system at that point, with an award which would presumably be called an ordinary degree

* to embark on the third year of a three-year course, leading to an honours degree

ÖйúAƬ

* to embark on the third year of a four-year course, leading to the award of a masters degree

Students who chose to embark on a masters course could still leave it at the end of their third year, with an honours degree. There would need to be some qualifying level which students would have to achieve in order to be allowed to transfer to the masters course; this might be that students would have had to obtain at least a 2.1 at the end of their second year. These proposals are phrased in terms of the degree structure in England and Wales. The considerations which give rise to them are equally valid in Scotland; but their implications for the Scottish system are not the same.

The third year of the honours course and the third year of a masters would be bound to have a good deal of material in common; in some universities they might even be identical. Logically, it should therefore be possible for a student who had obtained an honours degree (of good enough standard) to be able to proceed to the final year of a masters course. But it would be for individual universities to decide whether, and on what conditions, to allow such a transfer.

Almost certainly, some universities would have too few students wishing to take a master's course to make it economical to provide one - or to provide as big a range of options as masters students would expect. This means that provision must be made for such students to be able to transfer to another university at the end of their second year. But there should be no difficulty about this: most universities would welcome such students even if one or two did not.

ÖйúAƬ

Sir Peter Swinnerton-Dyer is professor in pure mathematics at the University of Cambridge.