scarce. Meanwhile the war raged, and there was no option but to take drafted men from all sections, Montreal in particular. Many could not speak intelligible English, and few had enjoyed any educational advantages. The men who came as cadets to be trained as pilots in 1918 graded much lower in personal and physical qualifications than the type of the previous year. And yet these same drafted men, who had withstood for three and a half years the call of their country, had more control over their machines at the end of their course than the men of the year before.

At the end of four, five, or six hours' solo these men could do all the high maneuvers, commonly thought dangerous, such as the barrel roll, the loop, the stall turn, the Immelmann turn. An astounding showing compared to the boys of 1917, who were forbidden to stunt and who rarely disobeyed the orders. In our American service we had specially selected men. They were college men, tested, qualified, and picked. But our men—and it's no reflection on them—seldom did their higher maneuvers with less than fifty hours of solo flying.

There is just one answer—it is a matter entirely of training.

It might be said that the Canadian casualties on the Texas flying-fields near Fort Worth during the winter of 1917-18, when the Royal Air Force occupied two airdromes, were the cause of comment all over the country. There were fifty fatalities in twenty weeks of flying, and machine after machine came down in a fatal spinning-nose dive, or tail spin, as the Americans speak of the spin.

Shortly after the Royal Air Force returned to its airdromes in Canada in the middle of April the Gosport system of flying training, which had been used successfully in England, was begun on the Curtiss J.N. 4B-type training-plane. The result was an immediate and material decrease in fatal accidents. In July, 1918, there was one fatality for every 1,760 hours of flying, and by October fatalities had been reduced to one in every 5,300 hours of flying. That is a remarkable achievement, as official data from other centers of training show one death in a flying accident for every 1,170 hours.

Briefly, the Gosport system is a graduated method of flying instruction. The cadet is led by easy steps through the earlier part of the training, and only after he has passed aerial tests in the simpler methods of control is he allowed to continue with the rest of his course and "go solo." The scheme provides that before he goes solo he must have spun, and shown that he can take his instructor out of a spin. Only then is he considered fit to go on his own.

"Dangerous" and "Safe" as terms to describe flying technique gave way to wrong and right. There was built up under sound instruction one of the best schools of flying in North America, the School of Special Flying, at Armour Heights, Ontario. There is no reason why there should not be established in this country a number of such schools, under men who have had army experience, to train great numbers of civilian fliers within the next few years. There is going to be a strong demand for the best flying instruction that can be given. It should be noted that only the most perfect system of flying instruction should be used, for the best is safest, and the safest, no matter how expensive, is comparatively cheap.

There is no reason why there should be an extended period of ground instruction for the non-military pilot of the future. He should be taught the elementary principles of the theory of flight, should know something about the engine with which he is going to fly, and understand some things about the rigging of his airplane. The details could come to him in constant association with the airplane before, during, and after each flight. No time need be spent on such subjects as artillery observation, machine-gunnery, wireless, bombing, photography, patrol work, and other subjects of a purely military nature, on which so much stress has been laid in training army pilots.

"What is an airplane?" Before going ahead with the method of Gosport instruction every pupil is given a lecture on the ground in which he is asked that question. One definition which was passed out to us in Canada was, "An airplane is a machine...." At this point the flight sergeant in charge of rigging would look dreamily into the distance. "An airplane is a machine...." he would begin again with an air of utter despondency. That was certainly no news to cadets. They had an idea that it might be a machine, and wanted to know more about it.

"An airplane is a machine with lift-generating surfaces attached to a frame which carries an engine, fuel, aviator, and devices by which he steers, balances, and controls his craft," the mournful flight sergeant was finally able to convince them.

Lift-generating surfaces—these are the bases of all flying. Every one knows, for instance, that a paper dart, instead of falling directly to the floor, sails in a gliding angle for some distance before crashing. Lift is generated under those plane surfaces moving through the air—and the lift keeps that paper dart gliding. Little eddies of air are compressed under its tiny wings. Imagine an engine in the dart, propelling it at some speed. Instead of having to nose down to get enough speed to generate lift under its wings, the dart would be able to fly on the level, or even climb a bit.

Just so with an airplane. A gliding airplane about to land with power shut off is that paper dart on a large scale. The airplane flying is the dart with power. To make the airplane safe to fly, to give control to the pilot so that he may steer it where he wants to, there is a rudder, moved by a rudder-bar under the foot of the pilot. It is impossible to turn a swiftly moving airplane in the air by the rudder alone. It must be banked to prevent skidding, even as a race-track is banked high on the turns. On its side an airplane will cushion its own bank of proper degree by the use of ailerons. These ailerons are sections of the wing-tips which may be moved either up or down. They are counterbalanced so that movement of the left down gives you the right aileron up. With left aileron down, the lift of the left wing is increased, and it tips up; at the same time the lift of the right wing is decreased, and it sags down. In that way the airplane is tipped up for a bank. These ailerons, wing sections, really, are controlled by a device known as the joy-stick in the cockpit.

We have seen how an airplane is made to tip and turn. Before a machine is under control we must be able to climb, or come down to the ground for a landing. Vertical control of an airplane is attained by the use of elevators, flaps on the tail plane acting as horizontal rudders. A pull-back on the joy-stick lifts the flaps, raises the nose of the machine, and causes it to gain height. Push the joy-stick forward, the elevators are turned down, and the machine goes into a dive for the ground. In making many maneuvers all three controls, rudder, ailerons, and elevators, are used at once and the pilot feels his way with the machine, guiding it with the stick and the rudder-bar.

After the explanation of the use of these controls, and their demonstration on the machine as it awaits its turn in the air, the pupil is taken up for his first ride—strictly a joy ride, and not always joyous for those who take every chance to be seasick. After he has a glimpse of what the ground looks like from the air, and has recovered from his first breathless sweep off the ground, the pupil is given a lesson in the demonstration of controls. The instructor explains through a speaking-tube attached to his helmet the very