AERIAL LOCOMOTION 245 



on the extremity of the wing, shows that tne surface 

 of the wing is differently inclined during the various 

 phases of movement. Now, we have seen that the two 

 limbs of the 8 described in this way are unequally 

 luminous, and from this we concluded that while one 

 of the limbs was being described, the inclination of 

 wing must have been more favourable for the reflection 

 of the sun by the gold spangle, and that, while the 

 other limb was being described, the gold spangle must 

 have had a less favourable inclination. If this is so, by 

 altering the position of the insect, a change should be 

 effected in the degree of reflection, and as a matter 

 of fact this is exactly what does happen. When the 



Fig. 173. — Experiment to test the direction of movement of an iDsect's wing. 



insect is turned through an angle of about 90°, what 

 was before the bright limb of the 8 is now no longer 

 conspicuous, whereas the other side, in its turn, becomes 

 brilliantly illuminated. In this way the variations in 

 inclination can be demonstrated. Even the angle 

 formed by the wing with the axis of progression could 

 be deduced from these experiments, but we shall see 

 that, by the employment of photography, this angle 

 varies from moment to moment. The most that one 

 can say is that the angle is about 45°, sometimes in one 

 direction, sometimes in the other. 



To determine with accuracy the direction taken by 



