IBS ANIMAL MKCIIANISM. 



besides, that during the alternate movements of flight, tl.c 

 plane of tlie wing changes its inclination with respect to the 

 axis of the insect's body, and that the upper surface of the 

 wing turns a little backward during the period of ascent, 

 whilst it is inclined forward a little during its descent. 



If we gild a large portion of the upper surface of a wasp's 

 wing, taking precautions that the gold-leaf should be limited 

 to this surface only, we see that the animal, placed in the sun's 

 rays, gives the figure of 8 with a very unequal intensity in 

 the two halves of the image, as represented in fig. 71. The 

 figure printed thus H gives an idea of the form which is then 

 produced, if we consider the thick stroke of this character as 

 corresponding with the more brilliant portion of the imago, 

 nnd the thin stroke as representing the part which is less 

 bright. 



It is evident that the c;;use of the phenomenon is to be 

 found in a change in the plane of the wing, and consequently 

 in the incidence of the solar rays; being favourable to their 

 reflection during the period of ascent, and unfavourable during 

 the descent. If we turn the animal round, so as to observe 

 the luminous figure in the opposite direction, the 8 will then 

 present the unequal splendour of its two halvis, but in the 

 inverse direction; it becomes bright in the portion before 

 relatively obscure, and vice versa. 



We shall find in the employment of the graphic method, 

 new proofs of changes in the plane of the wing during flight. 

 This phenomenon is of great importance, for in it wo seem to 

 find the proximate cause of the motive force which urges for- 

 ward the body of the insect. 



In order to verify the preceding experiments, and to assure 

 oui selves still more of the reality of the displacement of the 

 wing, which the optical method has revealed to us, we have 

 introduced the extremity of a small pointer into the interior of 

 the figure 8 described by the wing, and we have proved that 

 in the middle of these loops there really exist free spaces of 

 the form of a funnel, into which the pointer penetrates with- 

 out meeting the wing, whilst, if we try to pass the intersection 

 where the lines cross each other, the wing immediately strikes 

 against the pointer, and the flight is interrupted. 



