With some airplanes the approach to the stall is accompanied by changes in the behavior, such as tail buffeting or changes in the control characteristics of the airplane so that the pilot obtains a warning of the impending stall. Vith other airplanes it is possible to approach the stall without any perceptible warning other than the reading of the air-speed meter, in which case the danger of inadvertent stalling is considerably greater. Although it is not within the scope of this paper to discuss stalling characteristics, it is desired to point out that in general the danger of inadvertent stalling is greatest with those airplanes that behave worse when the stalling occurs; that is, with airplanes in which the stall starts at the wing tips. A warning of the impending stall is desirable in any case, but is particularly desirable with airplanes of the latter type.

"An investigation was made in the N.A.C.A. 7- by 10- foot wind tunnel to determine the aerodynamic section characteristics of an N. A. C. A. 23012 airfoil with a single main slotted flap equipped successively with auxiliary flaps of the plain, split, and slotted types. A test installation mas used in which an airfoil of 7-foot span was mounted vertically between the upper and the lower sides of the closed test section so that two-dimensional flow was approximated. On the basis of maximum lift coefficient, low drag at moderate and high lift coefficients, and high drag at high lift coefficients, the optimum combination of the arrangements was found to be the double slotted flap" (p. 1).

"Wing ducts for liquid-cooled engine radiators have been investigated in the N.A.C.A. full-scale wind tunnel on a large model airplane. The tests were made to determine the relative merits of several types of duct and radiator installations for an airplane of a particular design. In the test program the principal duct dimensions were systematically varied, and the results are therefore somewhat applicable to the general problems of wing duct design, although they should be considered as preliminary and only indicative of the inherent possibilities" (p. 1).

"Tests have been conducted in the NACA 8-foot high-speed wind tunnel to determine the effect of exposed rivet heads and spot welds on wing drag. Most of the tests were made with an airfoil of 5-foot chord. The air speed was varied from 80 to 500 miles per hour and the lift coefficient from 0 to 0.30" (p. 1).

"Tests were made in the NACA 8-foot high speed wind tunnel of a metal-covered, riveted, 'service' wing of average workmanship to determine the aerodynamic effects of the manufacturing irregularities incident to shop fabrication. The wing was of 5-foot chord and of NACA 23012 section and was tested in the low-lift range at speeds from 90 to 450 miles per hour corresponding to Reynolds numbers from 4,000,000 to 18,000,000. At a cruising condition the drag of the service wing was 46% higher than the drag of a smooth airfoil, whereas the drag of an accurately constructed airfoil having the same arrangement of 3/32-inch brazier-head rivets and lap joints showed a 29% increase" (p. 1).

Tests have been made in the NACA 8-foot high-speed wind tunnel of the drag caused by four types of lap joint. The tests were made on an airfoil of NACA 23012 section and 5-foot chord and covered in a range of speeds from 80 to 500 miles per hour and lift coefficients from 0 to 0.30. The increases in profile drag caused by representative arrangements of laps varied from 4 to 9%. When there were protruding rivet heads on the surface, the addition of laps increased the drag only slightly. Laps on the forward part of a wing increased the drag considerably more than those farther back.

"Tests have been made in the N.A.C.A. 8-foot high-speed wind tunnel of the drag caused by roughness on the surface of an airfoil of N.A.C.A. 23012 section and 5-foot chord. The tests were made at speeds from 80 to 500 miles per hour at lift coefficients from 0 to 0.30. For conditions corresponding to high-speed flight, the increase in the drag was 30 percent of the profile drag of the smooth airfoil for the roughness produced by spray painting and 63 percent for the roughness produced. by 0.0037-inch carborundum grains" (p. 1).

"Profile-drag coefficients published from tests in the N.A.C.A. variable-density tunnel (Technical Reports Nos. 460, 537, 586, and 610, references 1 to 4) have tended to appear high as compared with results from the N.A.C.A. full-scale tunnel (Technical Report No. 530, reference 5) and from foreign sources (references 6 to 8). Such discrepancies were considered in Technical Report No. 586, and corrections for turbulence and tip effects were derived that tended to reduce the profile-drag coefficients, particularly for the thicker airfoils. The corrected profile-drag coefficients, designated by the lower-case symbol cdo as contrasted with the older CDO, have been employed in the airfoil reports published since Technical Report No. 460, but even these corrected results continued to appear high, particularly for the thicker sections" (p. 1).

"An investigation was made in the NACA 7- by 10-foot wind tunnel of a large-chord wing model with a duct to house a simulated radiator suitable for a liquid-cooled engine. The duct was expanded to reduce the radiator losses, and the installation of the duct and radiator was made entirely within the wing to reduce form and interference drag. The tests were made using a two-dimensional flow set-up with a full-span duct and radiator" (p. 1).

"The systematic investigation of wing cooling ducts at the NACA laboratory has been continued with tests in the full-scale wind tunnel on ducts of finite span. These results extend the previous investigation on section characteristics of ducts to higher Reynolds numbers and indicate the losses due to the duct ends. The data include comparisons between ducts completely within the ring and the conventional underslung ducts" (p. 1).

"An investigation was made in the NACA 5-foot vertical wind tunnel of a large variety of duct inlets and outlets to obtain information relative to their design for the cooling or the ventilation systems on aircraft. Most of the tests were of openings in a flat plate but, in order to determine the best locations and the effects of interference, a few tests were made of openings in an airfoil. The best inlet location for a system not including a blower was found to be at the forward stagnation point; for one including a blower, the best location was found to be in the region of lowest total head, probably in the boundary layer near the trailing edge" (p. 1).

"The torsional deflection of the blades of three full-scale duralumin propellers operating under various loading conditions was measured by a light-beam method. Angular bending deflections were also obtained as an incidental part of the study. The deflection measurements showed that the usual present-day type of propeller blades twisted but a negligible amount under ordinary flight conditions. A maximum deflection of about 1/10th of a degree was found at V/nD of 0.3 and a smaller deflection at higher values of V/nD for the station at 0.70 radius" (p. 1).