Nice car: roomy, nippy and economical - but is it safe?

New crash tests designed for safety-conscious motorists yesterday provoked a row between the motor industry and consumer groups.

All seven small cars failed tests by the Transport Research Laboratory in Berkshire, provoking outrage from the Society of Motor Manufacturers and Traders, whose spokesman, Roger King, said: "These tests have been designed on purpose so that cars fail them. Cars are not built to these standards and therefore they cannot possibly meet them".

The pounds 1m testing programme ranked the Rover 100, the latest version of the Metro, which is one of Britain's best-selling small hatchbacks, as the worst for car safety in the results issued by the Department of Transport- sponsored crash tests. It received just one star out of a possible four in the Euro New Car Assessment Programme (NCAP) ratings. The Vauxhall Corsa, Nissan Micra, Fiat Punto and Renault Clio fared slightly better, scoring two stars.

The Ford Fiesta, Britain's best-selling small car, and the Volkswagen Polo topped the table with three stars.

The survey findings will be published in What Car? and be made available to all 12 million RAC and AA members. However, the Government will not force dealers to carry the information.

``We think and hope that industry will use these results to promote their cars," said John Bowis, Road Safety minister.

Two new cars of each model were bought by the testers and subjected to side and front-impact tests. Crashes with pedestrians were also tested by shooting dummy limbs and heads at the front of the car.

In the tests the Corsa, which was bought by nearly 76,000 motorists last year, saw the passenger's head severely damaged by hitting the glove compartment in a frontal collision. The driver's knees in a Renault Clio were found to be particularly vulnerable and many of the airbags did not make a clean contact with the dummies.

Motoring organisations welcomed the results. `` At last the consumer will be able to buy a safer car," said Edmund King, a spokesman for the RAC, one of the test sponsors. Research from Sweden has shown that half of fatal and disabling injuries could be saved if people chose the cars that gave the best protection in each weight category.

Rover disagreed with the findings, saying that the tests were "more severe than required by current legalisation". The company said the tests were "too fast". However Adrian Hobbs, of the Transport Research Laboratory, which carried out the tests, said ``we had to consider speeds where fatalities were likely to occur".

The manufacturers have criticised the tests, arguing that they carry out much more extensive testing on their models. However, Edmund King of the RAC pointed out that manufacturers have failed to release information to allow people to compare models on the basis of safety. He said: "This type of testing is carried out in Australia and is very influential in determining which models people buy."

One of the testers said he was very sceptical of the manufacturers' willingness to publicise a car's failings: ``We all see manufacturers shouting about how quickly they go from 0-60 mph. How willing would they be to say how slowly they can do 60-0 mph?" said one engineer.

Dealerships have also criticised the programme.``If you have a Rover dealership it is going to be very difficult to say `Oh yes, it is much less safe than the other make' ... I mean, it hardly helps," said David Leibling, a spokesman for Lex Service, one of Britain's largest motor showroom groups

Sheila McKechnie, the chief executive of the Consumers' Association, said that she would be writing to the Association of British Insurers to see whether the results could be used to lower premiums for certain marques.

The NCAP tests will also test family saloons, with the results being published later this year.

Car manufacturers have fought on the Continent to prevent the testing programme from receiving European funding. Martin Bangemann, the industry commissioner, wrote to safety organisations last year warning them that such testing could damage the "brand image" of European car-manufacturing companies. The rating system allows customers to see how likely they would suffer from serious injuries if they were driving in one of the cars tested. The worst results would mean that drivers would have a 20 per cent chance of incurring an injury, while the best would mean drivers had a one in 20 chance of escaping without a scratch.