On the morning of 21 May 1998, two cargo ships met in the Bay of Gibraltar. “Fedra,” loaded with bulk cement, was heading in to anchor at Algeciras. “Seafarer 1” was heading out. The bow of Fedra struck Seafarer 1’s starboard side at roughly a 60–70° angle. Fedra’s forepeak flooded and her bow sat lower in the water, but the crew and surveyors initially thought the damage was contained to the front of the ship. Days later, everything changed: on 30 May, the chief engineer found seawater pouring into Fedra’s No. 4 hold through a thin, badly rusted steel box built into the side—the discharge or “hat” box for the wing ballast tank. He plugged the opening with a wooden “fid,” but by then seawater had also reached No. 3 hold through corrosion holes in the bulkhead. The cement cargo was spoiled and the ship was treated as a constructive total loss.
Both owners blamed each other for the collision, and that liability was compromised at 50/50. But a key question remained for the Admiralty Court: did the collision cause the later flooding in the No. 4 hold, or did that flooding come from pre-existing unseaworthiness (extreme corrosion) that would have bitten sooner or later anyway? To keep this simple: the owners of Fedra (the defendants in this issue) had to prove, on the balance of probabilities, that the crash triggered the leak.
Everyone agreed on several things. The bottom plate of the discharge box was severely wasted—under 1 mm in places. If it opened up, seawater would flow into No. 4, then through existing holes into No. 3. A collision sends a shock wave through the hull that could disturb rust scale, especially where metal is pitted. So it was physically possible for a collision to knock off a fragile rust “scab” covering tiny perforations and start a leak. The fight was not about physics in the abstract. It was about timing and proof.
The defendants leaned on draft analysis: compare how deep the ship sat before and after the collision and over the following days; infer how much water was aboard; match that to an expected flow through a hole of a given size. Their expert tried to show a post-collision inflow of around 400 tons per day, consistent—he argued—with a small opening (he modeled about 32.5 mm). The problem was reliability. Many draft figures were photos taken in swell, rough estimates, or computer extrapolations. When the judge looked closely, several “anchor” numbers didn’t hold. Some credible figures, if accepted, implied there was already a large amount of water aboard right after the collision—too much to fit the defendants’ story that the leak began only because of the crash. Flow-rate math also strained: taking friction into account, the modeled daily inflow demanded a larger, rougher hole than the one posited.
Could the leak have started earlier—during the voyage from Greece—because the metal was already paper-thin? The judge didn’t find decisive proof of pre-collision ingress either; he simply found the defendants’ timetable not proven. The cement’s final condition couldn’t date the leak. The idea that a last-second helm movement (separate from impact) “triggered” the failure was speculative.
So the court came back to the legal test: who carried the burden and did they discharge it? The answer was no. The defendants did not prove, on a balance of probabilities, that the water entering No. 4 and the cargo damage were caused by the collision. Because causation failed, the court didn’t need to decide the fallback questions (like whether the collision would still count as the legal cause even if a rust scab made failure more likely). The judge added that if the timing had been proven—if the leak clearly started at or around the crash—he would have been inclined to treat the collision as the legal cause. But the facts didn’t get him there.
Final judgment, plainly stated: the preliminary issue was answered against Fedra’s owners. They failed to prove that the damage to No. 4 hold and its cargo was caused by the collision between Seafarer 1 and Fedra, so the collision was not established as the legal cause of that loss.
