Whirlwinds and tornadoes are rare phenomena, but if met with, it is worth while to take some trouble to put on record at least the hour of their appearance (local time), the direction in which the whirl moves onward, and the breadth of the path of destruction it leaves behind. When a storm of wind has passed over a wooded region and blown down many trees, the direction in which most of the trunks lie is worth observing. The top of the tree usually falls in the direction in which the wind was blowing, hence the root usually points to the direction of the wind. Waterspouts are closely allied to whirlwinds, and in any of those phenomena of revolving columns of air it is of much theoretical importance to determine the direction of the whirl about the axis, _i.e._, whether the rotation is in the direction of the hands of a watch or the opposite. The prevailing wind of a district may often be discovered by the slope of trees growing on open ground, or still better, by the difference in the degree of wave erosion on small lakes. If the banks are of the same material all round, the side against which the prevailing wind drives the waves will always be the most worn away. _Cloud and Sunshine._—It would be impossible to keep a record of the countless changes in the cloud-covering of an English sky, but in many parts of the world the absence or presence of cloud is a function of latitude, altitude, and season, of great stability, and worthy of being attentively studied. The amount of cloud is usually estimated as the number of tenths of the sky covered; but it is a very difficult thing indeed to compare a tenth of the visible sky near the horizon with a tenth near the zenith. There is no difficulty, however, in observing when the sky is completely overcast or quite free of cloud, and as a matter of convenience the belt round the horizon to the height of thirty degrees may be neglected, _i.e._, the lower third of the distance from the horizon to the zenith. Very often it will be found that clouds form and disappear at certain hours of the morning or evening, and it is useful to get exact information on the subject.