REMARKABLE parallels have emerged between the collision at Årsta in Norway on January 4, in which 19 people died, and the Ladbroke Grove collision near London Paddington on October 5 1999 which killed 31 (RG 11.99 p703).

In both cases, a train headed by a diesel locomotive met a DMU head-on at a closing speed in excess of 160 km/h, apparently because one of the drivers had ignored a red signal. Automatic train protection was fitted to the crucial signal in Britain, and to both trains in Norway, but matching equipment was absent. Large quantities of fuel were spilled on to the track, resulting in carriages being engulfed in flames as passengers struggled to escape from the wreckage.

Most of these factors are coincidental. Unlike aircraft, modern rail accidents seldom result in a fire. Train protection programmes were already under way in both countries that would have prevented these collisions. It would clearly be impractical to design fuel tanks capable of surviving accidents in which bogies are torn off.

The common thread that inevitably links these accidents with the Eschede disaster in Germany of June 2 1998 (RG 7.98 p449) is the structural disintegration of rolling stock formed from aluminium extrusions.

At Årsta and Ladbroke Grove, the leading car of each DMU suffered total destruction in a way that does not seem to occur with steel bodies, no matter how severely they may be deformed or crushed by an impact. It is apparent from photographs that much of the material forming the bodies of these leading cars had fragmented into quite small pieces; the trailing cars remained structurally intact, protected no doubt by the dissipation of kinetic energy in destroying the lead vehicle. Particularly striking is the way one bodyside survived complete with windows in each accident, as if to emphasise the extent to which the rest of the structure was unrecognisable.

Seam-welded aluminium extrusions are firmly established, and a large proportion of passenger stock on order in the European Union uses this form of construction. The destructive forces unleashed in all three accidents were quite exceptional. At Eschede, the train was travelling at 200 km/h when it derailed and struck a bridge, bringing one span down on top of the coaches. The closing speed at Årsta has been estimated at 160 to 180 km/h. No official estimate has been given for Ladbroke Grove, but 190 to 220 km/h is likely.

Given the commitment to aluminium extrusions, there has been understandable reluctance to accept visual evidence that this form of coach body may have a failure mode which causes splitting either along the welds or the webs of the extruded profiles. Clean ruler-straight edges that have appeared in numerous photographs are, to say the least, suggestive. Presumably, official reports into Eschede and the two more recent collisions will clinch the matter.

It will be surprising indeed if they do not call for more research into ways of improving the structural integrity of aluminium bodies. If the problem lies with the welded seams, it may be that bolts would help, particularly in joining steel and aluminium components. The issue of sacrificial crumple zones and crushable noses will surely be revisited.

That much said, it would be unfortunate if the highly emotional reaction to a 101 death toll at Eschede led to outright rejection of a lightweight and practical manufacturing technique in widespread use. ATP is spreading steadily across Europe, which must diminish the probability of high speed collisions still further.

Aircraft bodies can break up into small pieces in a similar way, but nobody suggests they should be made of steel. Instead, the emphasis is on preventing crashes, as a result of which air travel has displaced rail as the safest mode of travel. This lesson should guide rail safety regulators as well.