Reports that the International Cricket Council has called for a meeting with the MCC's world cricket committee about its proposals for a world Test championship raise the prospect that, finally, some shape is to be brought to the Test cricket calendar.

The image of Test cricket has suffered from the strong tournaments that the Twenty20 format has produced - two successful stagings of both the World Twenty20 competition and Indian Premier League. The five-day game's random and largely unintelligible programme looks outmoded.

Series are arranged at the whim of the national boards, whose commitment to the Future Tours Programme is lukewarm at best.

However, the MCC world cricket committee's plan - devised by Martin Crowe, the former New Zealand captain - looks itself to be unwieldy verging on the unworkable.

Crowe wants the championship to encompass all Tests in the calendar, with every side playing every other in a three-match series. This would threaten much of what is good in the existing schedule - the Boxing Day Tests at Melbourne and Durban, the New Year Tests at Sydney and Cape Town, the Lord's Test in July. Can these blue-chip fixtures be protected under such an all-consuming programme?

Also, it would take perhaps two years to play out an eight-team league in which each side plays 21 Tests. And what happens to the most watchable and valuable series such as the Ashes or Australia v India; these events surely demand series of more than three matches. To cut them down makes little sense.

The MCC plan is perhaps too ambitious. What would be better would be to retain something of the present system under which teams play series of varying length according to the status of the opposition. As happens now, the ICC's Test rankings could be drawn from this programme. What might then happen, I would suggest, is that a world Test championship is held every three years made up of quarter-finals, semi-finals and final, matches arranged according to placings in the rankings table - first v eighth, second v seventh, and so on.

For this tournament alone, Test matches would become limited-overs affairs, thus eliminating the draw and ensuring a definite winner. Each innings would be restricted to a maximum of 110 overs.

This may sound radical but in practice may not prove as radical as all that. Many Test matches finish naturally within these limits. Take the recent Lord's Test, which England won by 115 runs. There, the four innings occupied 101.4 overs, 63 overs, 71.2 overs (England declaring at six wickets down) and 107 overs. The game ended just before lunch on the fifth day.

Had Australia batted slightly longer in the fourth innings and their 110 overs elapsed, they would still have lost because they would scored fewer runs than England. What might well have changed is that England would have batted on longer in their second innings to make use of all their available overs, though they could too soon have been all out. No doubt critics would argue that ultra-defensive tactics might be employed but over innings of such length taking wickets would be the best way to keep down scoring.

Some games under this system might finish in what some would deem an unsatisfactory way but there are plenty of Test matches at present that are totally unsatisfactory - witness the recent Barbados Test in which bat dominated ball and a draw was clearly the only possible result from a very early stage. As a spectacle it was a travesty but under this plan batsmen-friendly pitches might still produce exciting games.

It might be worth an experiment, every bit as much as day-night Tests.