OCR
REFLEXIONS IN RIPPLED WATER 39 Plate XI shows the streak just before its disappearance and about as clearly defined as is possible in the open sea. In considering the formation of such a streak of light, we started by assuming the spectator to be looking stvazght across a succession of very regular waves, which, it was then shown, would cause a luminous point to appear in the water as a vertical line. But it might be argued that very regular waves travelling at right angles to the direction of sight would give a horizontal line in the water, and that therefore, if the sea were rough, with the sides of the waves facing in all directions, we should get, as the reflexion of a luminous point, no streak at all, but only a mere patch of light." That these cross waves are, however, of little account as compared with those " The statement on page 36, that the points a, 4, c, d, etc., in Fig. 14, all lie in the plane of the paper, is based on the supposition, made for the sake of argument, that the waves are perfectly regular, with the lines of their crests at right angles to the plane of the paper. But as a matter of fact, the waves are often very far from regular, and the motion of the water is generally exceedingly complicated, different wave-systems crossing each other in all directions, so that the normals to the surface are inclined in every possible direction within a certain angle from the vertical. The point P would under these circumstances be seen reflected at points on the surface of the water to the right and left of the vertical plane containing P and.Q, and the images thus formed will lie to the right and left of the vertical through P (Fig. 15), so that, as we have said, the combined effect of all the images will be a vague streak rather than a well-defined line. In order to meet the objection stated above, viz., that unless the movement of the waves towards or from the observer were much more marked than in other directions, the vertical streak would entirely disappear, let us consider the opposite extreme and suppose the waves to be moving across