348 



NA rURE 



[August 13, 190- 



is nearly the same as that of the constituent cell, the 

 successive stages are noteworthy. They sketch out in 

 a most interesting manner a reply to Newcomb's 

 criticism of the limits of application of the aeroplane 

 based upon the argument that increase of size means 

 diminished efficiency because, -for similar structures, 

 the weight varies as the cube while the area, upon 



The box kite of triangular section is, however, not 

 stiff as regards longitudinal shear, and the next " mile- 

 stone " marks the reduction of the triangular or pris- 

 matic form to the tetrahedron, an essentially stiff 

 framework for all directions. A tetrahedron of rods 

 with two adjacent faces covered with fabric forms a 

 tetrahedral kite cell which, on the principle of projec- 



FiG. I.— A Winged Tetrahedral Cell. 



which the lifting force depends, varies as the square 

 of the linear dimensions. 



The original stage, the ordinary kite, is a single 

 plane structure. The first step in advance is the 

 Hargrave box kite, with its upper and lower aero- 

 planes for its support, and side planes for stability. 

 ,To stiffen the framework of the box kite it must be 

 braced longitudinally and transversely; accordingly 

 Graham Bell's development commences by replacing 

 the rectangular framework of the box kite by a frame- 

 w;ork of triangular section which is by construction 



A Sixty-four celled Tetrahedial Kile. 



tion before referred to, is equivalent to three aero- 

 planes represented by the projections of the covered 

 sides upon planes at right angles. 



The further development of pure tetrahedral con- 

 struction is obvious. Four cells can be combined to 

 form a tetrahedron of double linear dimensions without 

 additional framework; the weight and wing area are 

 both simply proportional to the number of cells, and 

 not to the linear dimensions. For each set of four cells 

 thus combined there is an octahedral free space in the 

 interior which corresponds to the free space between 

 the two cells of the Hargrave kite. The tetrahedral 



Fig. 2.— a Four-celled Tetrahedral Kite. 



Stiff so far as the cross section is concerned. The 

 inclined sides are by the vector principle of resolution 

 of forces regarded as equivalent to their geometrical 

 projections, and, in so far as the principle applies, the 

 inclined faces represent the combined effect of aero- 

 planes of the area of the projections.^ 



1' This principle to be generally applicable would require the normal com- 

 ponent of wind pressure to be uniform and independent of the angle between 

 the plane and the wind. This is not the case with an aeroplane (see Rayleigh, 

 Nature, vol. xxv. p. io8) ; and for the principle to be applied approxi- 

 mately in the case of the kites some convention as regards the angle of 

 exposure of the aeroplanes to the wind would be required. 



NO. 1763, VOL. 68] 



kites that have the largest central spaces preserve their 

 equilibrium best in the air. 



Combining four multiple cells to fill the outline of a 

 tetrahedron of double size, again, we get a sixteen-cell 

 kite, and repeating the process again a sixty-four cell 

 kite, occupying a tetrahedron eight times the dimen- 

 sions of a single cell. The building up of multicellular 

 kites from the units is represented in the figures here 

 reproduced from illustrations in Dr. Bell's article. 

 Fig. I represents the unit cell, Fig. 2 a combination 

 of four cells. Fig. 3 of sixty-four cells. 



The kites fly with the points of the wings upward; 

 the line of junction of the covered faces of the tetra- 



