That is easy to understand. At the earth’s surface, for instance, light must fall (like all other objects) with a velocity equal to 981 centimetres at the end of a second. Now by the end of a second a luminous ray has travelled 300,000 kilometres. Suppose we could observe a horizontal luminous ray 300 kilometres long near the earth’s surface—a very far-fetched supposition—during the thousandth part of a second, which it will take the ray to pass from one observer to the other, it will fall to the extent of only about the five-thousandth of a millimetre. We can understand how it was that a luminous ray that deviates only to this imperceptible extent from its initial direction in the course of three hundred kilometres was always considered rectilinear. Is there no means of verifying whether light is or is not bent out of its path by gravitation? There is such a means in astronomy, as we shall now see. * * * * * It is impossible to detect the curvature of a luminous ray travelling from one point to another on the earth’s surface, mainly because weight on the earth is too slight to bend the ray much. A further reason is that our planet is so ridiculously small that we cannot follow the light over a sufficient distance. But what cannot be done on this little globule of ours, the entire diameter of which light can cover in the twenty-fifth of a second, may possibly be done in the gigantic laboratory of celestial space. We have, almost within our reach—a mere matter of 93,000,000 miles away, that is to say—a star on which weight is twenty-seven times greater than on the earth. We mean the sun. On the sun a body left to itself falls 132 metres in the first second. Its fall is twenty-seven times as rapid as on the earth. Hence, near the sun, light will be much more bent out of its path by gravitation. The deviation will be all the greater from the fact that the sun is 800,000 miles in diameter, and a luminous ray needs a much longer time to cover this distance than to travel the length of the earth’s diameter. Hence gravitation acts upon the ray of light during a much longer time than upon a ray that reaches the earth, and it will be all the more curved. Take a luminous ray that comes from a star at a great distance behind the sun. If it reaches us after passing near to the sun, it will behave like a projectile. Its path will no longer be rectilinear. It will be slightly curved toward the sun. In other words, the ray will deviate from a straight line, and the direction it has when our eyes receive it on the earth is a little different from the direction it had when it left the star. It has been diverted. Calculation shows that this deviation, though very slight, can be measured. It is equal to an angle of a second and three-quarters: an angle which the delicate methods of our astronomers are able to measure.