“Good. Then let us go a step further. Have you noticed the distribution of the numbers, letters and lines? The numbers are bunched, roughly speaking, towards the center, the letters round the edge, and the irregular lines between the two. Does this central mass give us anything?” “I get you,” Price replied. He had risen and begun to pace the room, but now he returned to the table and stood looking down at the photograph. “You know, as a matter of fact,” he went on slowly, “if, as you say, you take that central part which contains numbers only, the shape of the thing is not so very unlike England after all. Suppose the numbers represent land and the letters sea. Then this patch of letters in the top left-hand corner might be the Irish Sea, and this larger patch to the right the North Sea. And look, the letter circles form a band across the bottom. What price that for the English Channel?” French crossed the room, and taking a small atlas from a shelf, opened it at the map of England and laid it down beside the photograph. With a rising excitement all three compared them. Then Cheyne burst out irritably: “Confound the thing! It’s like it and it’s not like it. Let’s draw a line round those number circles and see if it makes anything like the shape.” He seized the photograph and took out a pencil. But just as in the scientific and industrial worlds discoveries and inventions seldom come singly, so among these three men the begetting of ideas begot more ideas. Scarcely had Cheyne spoken when French made a little gesture of comprehension. “I believe I have it at last,” he said quietly but with ill-concealed eagerness in his tones. “Those irregular lines in certain of the circles are broken bits of the coast line. See here, those two between 8 and U are surely the Wash, and that below H is Flamborough Head. Let’s see if we can locate correspondingly shaped outlines on the atlas, and fill in between those on the photograph with pencil.” A few seconds’ examination only were needed. Opposite, but slightly above the projection which French suggested as Flamborough Head was an angled line between GU and 31 which all three simultaneously pronounced St. Bee’s Head. Short double lines on each side of 24 showed two parts of the estuary of the Severn, and projections along the bottom near X and 27 were evidently St. Alban’s Head and Selsey Bill. That they were on the right track there could now no longer be any doubt, and they set themselves with renewed energy to the problem still remaining—the meaning of the circles and the numbers they contained. “We can’t locate the blessed things this way,” French pointed out. “We’ll have to rule squares on the atlas to correspond. Then we can pencil in the coast line accurately, and see just where the circles lie.” For a time measuring and the drawing of lines were the order of the day. And then at last the positions of the circles were located. They were all drawn round towns.