In craft built ages before steamers were designed, fishermen have observed that sails torn in the middle, if the rents were not too big, were more effective than when new and whole. What thus began in sheer wear, or accidental damage, is now imitated of set purpose. Under the equator one may often see small craft whose sails are matting woven with large openings, as the sailors say “to let out the wind.” The mariners of Carthegena, St. Thomas, and other islands of the West Indies, know that a ship goes better thus than if her sails were each one continuous breadth of canvas. Japanese junks of clipper builds have sails made of vertical breadths laced together so as to leave large apertures free to the air. Why is this breeziness of structure profitable? Because against the concave surface of an ordinary sail the wind rebounds so as to hinder its impulsive effect; through an aperture the air rushes in a continuous current and no rebound takes place. For a like reason, and with similar gain, Chinese rudders are made with separated boards or planks. The stream of water passing through such a rudder would exert an undesirable back pressure in a rudder of solid form. [Illustration: Perforated sails. 1, jib. 2, stay-sail. 3, square sail. 4, top sail. 5, sloop with perforated sails.] It would be interesting, and might prove gainful, to experiment with perforated sails in sail-boats, ice-boats and wind-mills. In large kites, sent to the upper air by meteorologists, it has been found helpful to give the fabric a few small perforations. Observations Must be Remembered and Compared: The Value of a New Eye. It is not only necessary to observe if one would learn, one must remember and compare observations. In a cycle of 223 lunations all the motions of the moon are repeated; it is astonishing that astronomers in Chaldea detected this period, exceeding eighteen years as it does. On the other hand, one of the most striking phenomena of a solar eclipse, its revelation of the solar corona, does not seem to have been noticed until comparatively recent times. The first known record of it is by Lobatchevsky, July 8, 1842.