In November, 1893, I launched several of these machines from the balcony of the tower of Boston Light, and more recently I have experimented from the top of the cliffs at Manomet. The former place is an ideal one for the purpose of experiment, being as it is, one hundred and eleven feet above the sea with a straight drop of seventy or eighty feet. Unfortunately, a gale of wind was blowing when I visited the light, and two out of the three machines were total failures, being badly bent by the wind before they were launched. The third machine righted itself before reaching the ground, but the pendulum, which will presently be described, was too light to do efficient work. The experiments from the cliffs at Manomet were even less successful, owing to the fact that the descent is not sheer. All of the machines failed to gain sufficient velocity to clear the cliff. Those who wish to experiment with machines weighing only a few pounds will probably find that a height of seventy or eighty feet will be sufficient if the position gives a straight drop. When it comes to experimenting with a soaring machine as large as Lilienthal’s and carrying a weight representing that of a man, the summit of Mt. Willard, near the Crawford House, N.H., will be found an excellent place. [Illustration: A Soaring Machine. An instrument for making scientific experiments Designed by James Means. _FIG 3. PLAN_ _FIG 4. SIDE ELEVATION_] To any one who desires to take up this most fascinating study, Figs. 3 and 4 will give a general idea as to the construction of his first instrument for making experiments. A represents a backbone five-eighths of an inch square and four feet long, made of pine wood; B, the main aeroplane, eight inches wide and three feet long. This should be made of light tin plate, and bent in the middle so as to form a flattened V; the angle should be about one hundred and seventy degrees. C represents a steering aeroplane six inches by twenty-four inches, pivoted at cc, also made of light tin plate; D, a vertical aeroplane four inches by twenty inches, rigidly fixed in the wooden backbone; E, a rod of steel wire, eighteen or twenty inches long, and carrying an adjustable leaden weight of three ounces; K, a rod two and one-half inches long, soldered in the centre of and vertical to the plane C, with a pivot at the upper end with which the rod MM is connected. This rod should have five or six pivot-holes at its forward end N, so that its working length may be varied for different experiments; J, a rod pivoted at G, free to swing fore and aft; N, a pivot where the rod MM joins the rod J; F, a leaden weight adjustable higher or lower upon the rod J; its proper weight is x, an unknown quantity. Upon ascertaining by repeated experiment the right _weight_ for F, the right _position_ for the adjustable weight E, and the right _length_ for the rod MM, the reaching of the maximum efficiency of a system of aeroplanes largely depends. I think that this