AERIAL LOCOMOTION 241 



doing left an interrupted track. If the wing had been 

 forced harder against the cylinder, the mark left would 

 have been somewhat in the form of a comma, and the 

 frequency naturally somewhat less owing to the re- 

 sistance due to friction. The same effect may be 

 observed in the movements of all kinds of animals. 



To calculate exactly the frequency of the wing 

 movements, the vibrations of a tuning-fork are simul- 

 taneously recorded on the cylinder. These vibrations 

 leave on the blackened paper an undulating line ; each 

 vibration representing ^L of a second. It now only 

 remains to count the number of marks traced by the 

 insect's wing on a length of paper corresponding to 

 250 vibrations of the tuning-fork. The number of 

 wing movements per second is thus obtained. 



By this method it was calculated that in the common 

 fly there were 330 strokes per second, in the bee 190, 

 and in the macroglossus of cheese rennet 72. Thus 

 obeying the general law applicable to birds, namely, 

 the smaller the species the more rapid the movements 

 of the wings. 



Synchronous Movements of the Wing's, Variations in 

 Surface Inclination. — There are other facts to be learnt 

 from direct registration of the wing movements. Thus 

 by holding a fly in such a position that its two wings 

 strike the cylinder at the same time, it will be seen 

 that both wings impart the same number of strokes, 

 and that the movements are absolutely synchronous. 



In some species of insects the upper surface of the 

 wings is covered with fluffy hairs, while the lower 

 surface is bare ; tracings taken from the wings of such 

 insects show alternate variations. Fig. 169 was 

 obtained from Macroglossus of cheese rennet, a small 

 diurnal hawk moth which flies very rapidly and is very 

 common in France. The insect was held in such a way 

 that the under side of its wing touched the cylinder. 



