Other necessary improvements are of minor urgency and in some cases will be easy to accomplish. Better mechanical construction, especially in the details of attachments, needs only persistence and common sense. Structural strength will be increased; the wide spread of wing presents difficulties here, which may be solved either by increasing the number of superimposed surfaces, as in triplanes, or in some other manner. Greater carrying capacity--two men instead of one--may be insisted upon: and this leads to the difficult question of motor weights. The revolving air-cooled motor may offer further possibilities: the two-cycle idea will help if a short radius of action is permissible: but a weight of less than two pounds to the horse-power seems to imply, almost essentially, a lack of ruggedness and surety of operation. A promising field for investigation is in the direction of increasing propeller efficiencies. If such an increase can be effected, the whole of the power difficulty will be greatly simplified. [Illustration: MELVIN VANIMAN TRIPLANE] [Illustration: JEAN DE CRAWHEZ TRIPLANE] [Illustration: A TRIPLANE] This same motor question controls the proposal for increased speed. The use of a reserve motor would again increase weights; though not necessarily in proportion to the aggregate engine capacity. Perhaps something may be accomplished with a gasoline turbine, when one is developed. In any case, no sudden increase in speeds seems to be probable; any further lightening of motors must be undertaken with deliberation and science. If much higher maximum speeds are attained, there will be an opportunity to vary the speed to suit the requirements. Then clutches, gears, brakes, and speed-changing devices of various sorts will become necessary, and the problem of weights of journal bearings--already no small matter--will be made still more serious. And with variable speed must probably come variable sail area--in preference to tilting--so that the fabric must be reefed on its frame. Certainly two men, it would seem, will be needed! Better methods for starting are required. The hydroplane idea promises much in this respect. With a better understanding and control of the conditions associated with successful and safe descent--perhaps with improved appliances therefor--the problem of ascent will also be partly solved. If such result can be achieved, these measures of control must be made automatic.