Inside we met the cox, the only full-time, paid member of the team, who is apparently widely respected throughout the region for his knowledge of the local waters and tides. The rest of the team is made up of volunteers, reachable by pager.

The cox was a little gruff at first, living up to his sea-dog credentials. The slant of his eyebrows indicated that he thought our search was hopeless. However, he mellowed as we explained our quest and pulled out his charts and tide almanacs.

The high tides which could have deposited the bottle on a Sunday in March 2002 were as follows:

3 March 04.11

10 March 11.38

17 March 03.16

24 March 09.47

According to the cox, the bottle would have needed a north-easterly wind to bring it into Warden Bay. It could have been thrown from a boat, a yacht or a cross-Channel ferry, although he pointed out that there were few pleasure boats around that area in the winter months. He thought it might well have been thrown from an English beach – Norfolk was his preference – but since the Evian bottle had not been commercially available in the UK, I was doubtful of this. The cox pointed out that if you throw a bottle out on an incoming tide it might well just come straight back at you a few waves later, but it seemed to us very unlikely that a Frenchwoman would have brought a bottle available only on the continent to such a bleak, isolated beach in England with such inconvenient access. Thrown from a ferry, he thought it could have taken days or at most weeks to wash up.

Entering into the spirit of the search, he then checked the tides for the coast of northern France. Thrown from Brest, Cherbourg or the Breton ports, the bottle would have gone west towards America and Spain, taken the Gulf Stream down the Portuguese coast or spun in a never-ending circle somewhere off St Malo. From Dieppe eastwards, and from the Oostende region, it could have drifted towards the Isle of Sheppey.

We stared gloomily at his charts, trying to tease out an easy answer – or any answer – but in vain.

I believed what the lifeboat skipper had told me, but keen sailor friends suggested that I should double-check with the Royal Navy who had virtually invented tidal almanacs. So, just in case there was any loophole of information that the tidal patterns could supply, I also checked with the Hydrographic Office, part of the Ministry of Defence, who supply the information on which the tide tables are based. The man who answered my query informed me that for his training he had studied the tidal flow of oranges – good buoyancy, bright colour, not much wind resistance, eventually biodegradable – which could be used to predict tides to within a few minutes of accuracy. But his email reply suggested that the bottle’s launching point was not so easy to track backwards: