Now, as to their connection. Of direct evidence we have little. A few meteors have been observed to come from the known radiants of shooting-stars. Two instances we have of the fall of meteorites during star showers. One in 1095, when the Saxon Chronicle tells us stars fell “so thickly that no man could count them, one of which struck the ground and when a bystander cast water upon it steam was raised with a great noise of boiling.” The second case was the fall of a siderite, eight pounds’ worth of nickel-iron, at Mazapil during the Andromede shower of 1885, which was by many supposed to be a part of the lost Biela comet. It contained graphite enough to pencil its own history, but unfortunately could not write. The direction from which it came was not recorded, and so the connection between it and the comet not made out. [Illustration: THE RADIANT OF A METEORIC SHOWER, SHOWING ALSO THE PATHS OF THREE METEORS WHICH DO NOT BELONG TO THIS SHOWER—AFTER DENNING.] If our direct knowledge is thus scanty, reasoning affords surer ground for belief. For at this point there steps in a bit of news about the family relations of shooting-stars from a source hardly to have been anticipated. Indeed, it arose from the thought to examine a qualitative statement in Young’s “Astronomy” quantitatively. Mathematics is simply precise reasoning, applied usually to the discovery that a pet theory will not work. But sometimes it presents one with an unexpected find. This is what it did here. It is an interesting fact of observation that more meteors are visible at six o’clock in the morning than at six o’clock at night in the proportion of 3 to 1. This seeming preference for early rising is due to no matutinality on the part of the meteors, but to the matin aspect then presented by the Earth combined with its orbital motion round the Sun. For at six in the morning the observer stands on the advancing side of the Earth, at the bow of the airship; at six at night he is at the stern. He, therefore, runs into the meteors at sunrise and slips away from them at sunset. He is pelted in the morning in consequence. Just as a pedestrian facing a storm gets wetter in front than behind. [Illustration: METEORS Diagram explaining their proportionate visibility. —————— _denotes true paths._ —— - —— ” _apparent paths._ —————— - - - ” _Earth’s path._] So far the books. Now let us examine this quantitatively according to the direction in which the meteors themselves may be moving before the encounter. Suppose, in the first place, that they were travelling in every possible direction, with the average velocity of the most erratic members of the family, the great comets. On this supposition calculation shows that we ought to meet 5.8 times as many at six in the morning as at six at night. If their orbits were smaller than this, say, something like those of the asteroids, we should find 7.6 to 1 for the ratio.