A great many changes in the weather take place after three in the morning. Most campers are accustomed to waking anyway once or twice to replenish the fire, and a glance at the stars will show the sleepiest what changes are occurring in the eternal panorama. A man may have gone to bed in security to get up in a snowstorm, whereas a survey of the skies at three would have noted the coming change. The habit of waking in the dead of night,--which isn't really so dead after all,--is not an unpleasant one. Its compensations are set forth in a beautiful and vivid chapter of Stewart Edward White's "The Forest." Every camper knows them, and this added mastery that a knowledge of the skies gives him lends a sense of power, which lasts until the unexpected happens. For the unexpected happens to the best regulated of all forecasters, the Government. Equipped with every instrument and with an army whose business is nothing else than to hunt down storms and warn the public, the Weather Bureau is still surprised fifteen times out of a hundred by unforeseeable changes in atmospheric pressure. It is scarcely likely then that amateurs without flawless barometers and without reports of the current weather in three hundred places could hope to foretell with complete accuracy. But there is a place for the amateur, aside from his own personal gratification and profit. The Weather Bureau within the limits of the present appropriation cannot expect to predict for every village and borough. That the amateur may do and with as great accuracy for the few hours immediately in advance. The Weather Bureau may predict with this large percentage of accuracy--85%--for forty-eight hours in advance because its scope is country wide. It may even forecast in a general way for seven days and still maintain a considerable advantage over almanac guesswork. But the man who is relying upon local signs is limited to ten or at most twelve hours. Of course he may guess beyond that but it is only a guess. The work that the Bureau does and that he may do within his limits is not guesswork. Meteorology is an exact science, and forecasting is an art. Both may be studied now in classes under professors with degrees in the same way that any other science and art may be studied. The old sort of weather wisdom which was a startling compound of wisdom, superstition, and inanity has passed away, or is passing away as rational weather talk spreads. These limits of the layman--ten hours with no instruments--are further defined by his locality. In mountainous country changes come more quickly than in level localities, in winter than in summer, so that one's prophetic time-limit is shortened.