I was using all my faculties in setting and keeping to the prescribed course. The Baker navigating machine, with the chart, was on my knees. Not knowing what kind of weather was before us, I knelt on my seat and made haste to take observations on the sea, the horizon, and the sun, through intervals in the covering of clouds. The navigation of aircraft, in its present stage, is distinctly more difficult than the navigation of seacraft. The speed at which they travel and the influence of the wind introduce problems which are not easily solved. A ship's navigator knows to a small fraction of a mile the set of any ocean current, and from the known speed of his vessel he can keep "dead reckoning" with an accuracy that is nearly absolute. In fact, navigators have taken their craft across the Atlantic without once having seen the sun or stars, and yet, at the end of the journey, been within five miles of the desired destination. But in the air the currents either cannot be, or have not yet been, charted, and his allowance for the drift resulting from them must be obtained by direct observation on the surface of the ocean. By the same means his actual speed over the ocean may be calculated. He finds the position of his craft by measuring the angle which either the sun or a selected star makes with the horizon, and noting the Greenwich mean time at which the observation is made. If the bearings of two distinct wireless stations can be taken, it is also possible to find his definite position by means of directional wireless telegraphy. When making my plans for the transatlantic flight I considered very carefully all the possibilities, and decided to rely solely upon observations of the sun and stars and upon "dead reckoning," in preference to using directional wireless, as I was uncertain at that time whether or not the directional wireless system was sufficiently reliable. My sextant was of the ordinary marine type, but it had a more heavily engraved scale than is usual, so as to make easier the reading of it amid the vibration of the aëroplane. My main chart was on the Mercator projection, and I had a special transparent chart which could be moved above it, and upon which were drawn the Sumner circles for all times of the day. I carried a similar special chart for use at night, giving the Sumner circles for six chosen stars. To measure the drift I had a six-inch Drift-Bearing plate, which also permitted me to measure the ground speed, with the help of a stopwatch. In addition, I had an Appleyard Course and Distance Calculator, and Traverse tables for the calculation of "dead reckoning." [Illustration: IT WAS HARD TO FIND AN AËRODROME WITH SUFFICIENT "TAKE OFF"] [Illustration: SIGHTSEERS, IF LEFT TO THEMSELVES, WOULD HAVE WRECKED THE MACHINE]