The first meteorograph, a combined recording thermometer and barometer (from which the height can be calculated), was constructed by Mr. Fergusson in August 1895, and three months later he united a recording anemometer to the thermometer, which was probably the first apparatus of this kind to be attached to kites. A meteorograph, recording the atmospheric pressure, air temperature, and relative humidity, was ordered from M. Richard of Paris in 1895, like one already carried by French aeronauts, except that, since for kites lightness is all-essential, M. Richard constructed this triple-recorder for the first time of aluminium, and hereby reduced its weight to 24⁄5 lbs. One of these meteorographs was hung to a ring at the point of attachment of the two kite-lines to the main line, a method which was used until recently. In August 1895, besides the Eddy kites, there was first used the cellular or box kite, invented by Lawrence Hargrave of Sydney, Australia, which bears no resemblance to the conventional forms of kites and which it would not be supposed could fly. As seen from [Fig. 9] its appearance is that of two light boxes without tops or bottoms, fastened some distance above each other. The wind exerts its lifting force chiefly upon the front and rear sides of the top box, the lower box, which inclines to the rear, and so receives less pressure, preserving the balance. The ends of the boxes, being in line with the wind, keep the kite steady and serve the purpose of the dihedral angle in the Malay kite. The Japanese are said to fly a single box, which is the prototype of the Hargrave double cell.
Fig. 9.—Hargrave Kite.
At the present time some form of the Hargrave kite is generally employed for scientific purposes. On account of the weight of the large cord necessary to control these kites, and the surface which it presented to the wind, a height of 2000 feet could not be reached, so, during the winter of 1895-6, following Archibald's example and the methods of deep-sea sounding employed by Captain Sigsbee, U. S. N., steel pianoforte wire was substituted for the cord. This wire is less than half as heavy, and less than one-fourth the size of cord having the same strength, and, moreover, its surface is polished, which reduces the friction of the wind blowing past it. With the wire the height of a mile was reached in July, and a mile and two-thirds above Blue Hill in October 1896.
Up to this time a reel turned by two men sufficed to draw down the kites, but the increasing pull and length of wire made recourse to steam-power necessary. In January 1897 a grant of money was allotted from the Hodgkins Fund of the Smithsonian Institution for the purpose of obtaining meteorological records at heights exceeding ten thousand feet, and no doubt the first application of steam to kite-flying was the winch built by Mr. Fergusson with ingenious devices for distributing, oiling, and measuring the length of wire. The cumulative pressure of the successive coils of wire finally crushed the drum, and the next apparatus applied the principle of Sir William Thomson's deep-sea sounding apparatus, in which there is no accumulation of pressure. In October 1897 records were brought down from eleven thousand feet, or a thousand feet above the prescribed height.
The kites and apparatus at present employed at Blue Hill will now be described.
The kites are all of the multiplane type, and mostly of Hargrave's construction with two rectangular cells. These cells are covered with cloth or silk, except at their tops and bottoms, and one is secured above the other by four or more sticks. The wooden frames are as light as possible, but are made rigid by guys of steel wire that bind them in all directions. The average weight is about two ounces a square foot of lifting surface, which is about the same weight a square foot as the Eddy kites when all the surface is included in the estimate. The largest of the Hargrave kites stands nine feet high, weighs eleven pounds, and contains ninety square feet of lifting surface, which in the recent kites is arched, resembling the curvature of a bird's wings, a construction that was proposed many years ago by Phillips ([Fig. 10]). These curved surfaces increase the lift, or upward pull, more than the drift, or motion to leeward, and so the angular elevation is augmented without materially adding to the total pull on the wire, which should not exceed one-half its breaking strength.
