Loughor proved to be a straggling village situated on the left bank of the estuary where the latter made a right-angled bend towards the north. The two bridges ran side by side and a couple of hundred yards apart. That carrying the road was a fine wide structure of ferro-concrete, fairly new and leading directly into the village. The railway bridge was lower downstream, considerably older, and supported on timber piles. Both were about three hundred yards long, and built with short spans and many piers. The tide was out and the usual wide mudbanks were exposed on either shore. Directly French saw the spot he felt that here indeed was what he sought. On a dark night it would have been easy to drop the crate from the road bridge in absolute secrecy. Nor, as far as he could see from the map, was there any other place from which it could have been done. He had assumed that the criminal would select an ebb tide for his attempt, in order to ensure the crate being carried as far as possible out to sea. For the same reason French believed he would choose the time of its most rapid run. That time must also be in the dead of night to minimise the risk of discovery from passing road traffic. From 2 to 4 A.M. would probably best meet the conditions, as the chances were a thousand to one that the road would then be deserted. French wondered if he could get anything from these considerations. He turned to Manners. “I suppose it takes a bit of time to get up a good run in an estuary like this? How soon after high water would you say the current was running at full speed under the bridge?” “From one to two hours, more or less.” One to two hours previous to the period 2 to 4 A.M. meant between midnight and 3 A.M. “Now, Mr. Manners, can you tell me whether high water fell between twelve and three on any night about five or six weeks ago?” Manners once more produced his tide table. “Five or six weeks ago,” he repeated, slowly. “That would be between the sixteenth and the twenty-third of August.” He ran his stubby finger up the pages, then read out: “‘Twenty-first, Sunday, O point five—that’s five minutes past midnight, you understand. Twenty-second, Monday, one-twenty-three A.M.; twenty-third, Tuesday, two-fifty-five A.M.’ ’Ow would that suit you, sir?” “All right, I think,” French answered as he noted the three dates. “Any of those top springs?” “No, sir; you don’t get ’igh water of springs at night. ’Bout six or seven o’clock it runs. Those dates wot I gave you are about dead neaps.” “But there is still a strong flow at neaps?” “Oh, bless you, yes! Not so strong as at springs, o’ course, but plenty strong enough.” All this seemed satisfactory to French and he felt a growing conviction that the small hours of the 21st, 22nd, or 23rd of August had witnessed the launch of the crate. But this was mere theory, and theory is popularly admitted to be worth only one-sixteenth of the value of practice. Could not he arrive at something more definite?