FLIGHT OF THE DRAGON-FLY. 275 



There is a peculiarity in the flight of the Dragon-fly — at all 

 events in some species — which assists it greatly in chasing prey 

 It has the power of suddenly reversing the stroke of its Avings, 

 so that it can stop itself in full career, and can even fly back- 

 wards with some spee;fl. This power is specially useful when 

 the Dragon-fly is engaged in chasing an insect among obstacles, 

 and, when it has darted into a narrow passage in which there 

 is not sufficient space for it to turn, it can back itself out again 

 without much difficulty. 



It is a pity that the lovely colours with which most of these 

 insects are decorated should be so fleeting, but it is a sad fact 

 that there are scarcely any insects in which the colours last for 

 so short a time after death. The only plan for preserving 

 them that seems at all feasible is that which was employed by 

 the late Mr. Waterton, namely, taking the insect to pieces, 

 removinp; the interior, rtaintino- the inside of the emr>ty shell 

 with the requisite colours, and then putting the pieces to- 

 gether again. Still, even this tedious and laborious plan does 

 not answer with the comparatively solid thorax, and, as many 

 species have stripes of bright scarlet, yellow, blue or green on 

 the thorax, the process is evidently an imperfect one. The 

 species which is shown in the Plate, Lihellula depressa, 

 keeps its colour better than any which I know, but then its 

 hues are never at any time brilliant. Some of them retain 

 sufficient colour to show what their hues have been, but in all 

 cases the brilliancy of the various tints passes away and cannot 

 be renewed. 



We will now take in detail one or two of the most prominent 

 species as examples of this beautiful family. They are divided 

 into two large groups, called Libellulides and Agrionides. The 

 insects of these groups may easily be distinguished by the 

 shape of the head, which in the Libellulides is rounded, and in 

 the Agrionides is more cylindrical and set cross-wise to the 

 body, something like that of the hammer-headed shark. So 

 strongly is this mode of structure defined in some of the 

 species, that without their wings they bear no little resem- 

 blance to croquet-mallets. We will take two examples of each 

 group. 



On Woodcut XXVII. Fig. 2 is represented one of our finest 



T 2 



