The news that more time is to be made available for practice of numeracy and literacy in primary schools will be welcomed by many teachers and most parents.

If children are not competent in these areas, they will make little sense of most of the rest of the curriculum or, perhaps, the rest of their lives.

However, it is not only the low achievers who need help in basic skills. We also need mathematically able young people so they can become the skilled technicians and researchers of the future. In our preoccupation with numeracy we must not lose sight of the fact that many, indeed most, of our children need a solid grounding in mathematical logic if they are to master key topics, particularly algebra, in later years.

We need mathematically able primary teachers. Currently, the minimum qualification to enter training as a primary teacher is a grade C at GCSE. For more than half of our entry at Exeter University this is their highest maths qualification.

Compare this with mathematically high-performing countries such as Hungary and Poland. Not only is the mathematical attainment of their pupils at 16- plus higher than that of pupils in this country, but their prospective primary teacher trainees will also have studied mathematics for a further two years under a baccalaureate system. Our equivalent entry qualification, grade C at GCSE, is the mark obtained by more than 20 per cent of all candidates taking mathematics. Moreover, the difference in standard between a top grade C and a bottom grade C is considerable. It is also possible, indeed now usual, to gain a C on the intermediate tier of exam papers. There is little algebra on these exam papers and what there is can in effect be ignored and the required C gained by concentrating on “softer” topics such as handling data.

Teachers need to have a level of competence and understanding well beyond the level at which they are teaching. This is particularly true if, as is recommended by the Numeracy Task Force, we emphasise whole-class teaching rather than individual or group work.

Furthermore, high level skills are just as important for teaching the early primary years, for it is here that the foundation of mathematical logic must be laid, and by teachers who know what they are doing. It may not seem important that, for example, writing out the calculation of: (witnessed in a primary class recently) is wrong - after all, the pupil achieved the correct answer of 20 - but it is important that that pupil had not understood about balancing equations.

A teacher who understood maths would have insisted that the solution be written out correctly as: No wonder that our children have great difficulty in mastering algebra when one of the fundamental building blocks, that equations must balance, is not taught at a young age - and may never be set right.

In my opinion, the minimum mathematical qualification for intending primary teacher trainees should be a B in GCSE, even moving to an A in a few years. Unfortunately, there is little support for this proposal. Instead, the Teacher Training Agency has developed a mathematics curriculum for trainees. It certainly looks rigorous enough - but it is going to be policed by Office for Standards in Education inspectors. Alas, I doubt the ability of OFSTED inspectors to understand what is really happening throughout such a course. It is all too easy to put on a show for a week.

So how about a different approach? Let’s institute an externally set and marked national assessment of competence in mathematics for primary teachers, which could be taken even before the course starts (it could eventually be computerised and taken at any time) and which must be passed before Qualified Teacher Status is awarded.

In this way, we would ensure that trainers and trainees take mathematical competence seriously. The assessment need not be complicated - short tests of competence in mathematics, using questions such as those above, would suffice.

We must be confident that our primary teachers have the skills and knowledge necessary to enhance our pupils’ attainment well beyond the Government’s target of 75 per cent of pupils gaining level 4 at key stage 2. This can be no more than a starting point in developing a numerically and mathematically able nation for the future.

David Burghes is director of the Centre for Innovation in Mathematics Teaching at Exeter University