464 Dr Hankin, The. soaring flight of dragon-flies 



them. They kept at a distance of 1 to 3 metres above it. 

 The grass was waving and showed, where the dragon-flies 

 were ghding, no sign of its being sheltered by the tree. Some- 

 times, in a stronger gust, a few of the dragon-flies came up 

 to leeward of one of the branches. Rarely a few went further 

 to leeward than usual. The group generally extended from 

 near the tree to about 35 metres to leeward of it. The dragon- 

 flies showed lateral swaying, and one of them showed lateral 

 instability for an instant just after a gust. 



10.20. When quite near the tree the dragon-flies occa- 

 sionally flapped. Away from the tree they appeared to glide 

 only. Glides of at least 10 metres up wind without a flap 

 were observed. 

 In view of the probable speed of the wind on this occasion, it 



is likely that the speed of these dragon-flies through the air was 



above rather than below 15 metres per second. 



When in low speed flight dragon-flies appear to travel a little 



faster than a locust. The speed of locusts, whose flight depends 



on flapping, has been measured by me on different occasions and 



found to be 4 metres per second. 



VII. The relation of ascending currents to the soaring 

 flight of dragon-flies. 



The facts of the case are simple. Whenever the air is soarable 

 or, in other words, whenever there is sunshine, dragon-flies avoid 

 such currents. 



For instance, on one occasion, observations were being made 

 at a time when the sky was partially cJouded with short intervals 

 of sunshine. Dragon-flies of two species were gliding, so long as 

 there was cloud shadow, in the current reflected up from the 

 windward side of a stable. Whenever there was slight sunshine 

 the dragon-flies elevated their abdomens and glided away, generally 

 only for a short distance from the ascending current. But twice 

 during the period of observation the sun came out so strongly that 

 the dragon-flies put on a brake, that is to say they were gliding 

 with abdomens depressed. On each of these occasions every 

 dragon-fly went a long distance, 50 yards or more, from the 

 ascending current. 



Dragon-flies appear to remember the position of an ascending 

 current. At Futteypur-Sikri they have been seen collecting on the 

 windward side of the hill towards sunset, coming from distances 

 of half a mile or more to get there. 



Thus observations on dragon-flies yield valuable evidence that 

 soaring flight is not due to the use of ascending currents. As 

 happens with soaring birds, such currents are avoided so long as 

 the air is capable of supporting soaring flight. 



