It comes to pass in the following manner:--Some movements of a struggling kind take place inside the pupa case; and at length the case splits at the upper part, near the head: through this rent the body of the dragon-fly appears and tears it open, acting like a wedge, until the slit extends along the head across to the two eyes. The latter transverse slit is produced by a contrivance similar to that mentioned in the case of the blow-fly, a sort of air bladder which the insect distends at its pleasure, and thus causes the skin covering its head and eyes to split open. The head and body of the insect rise and make their appearance through the slit; and the head is now so much larger than it appeared while in the pupa case, that it seems almost impossible that it could ever have been contained within it. The insect continues to rise perpendicularly out of the case, and the legs make their exit, leaving the leg cases of the pupa undisturbed in their attachment to the support on which it rests. In order to facilitate the disengagement of the rest of its body the insect now bends itself in a curved form backwards, being only kept from falling by the last rings or segments of its body being still embraced by the sheath of the pupa case. When it has extricated itself thus far, it begins to move about its legs in different directions for several minutes, as if to get them into use; but after this it ceases all movement whatever. Not a quiver can be seen in its limbs, and the young observer would be inclined to conclude,--as a great entomologist once did, the first time he witnessed the changes of this insect,--that it was in reality dead, and that it was a waste of time to watch it any longer. This state of profound inaction may endure for a quarter, or even half an hour: it appears to be intended in order to give time to the insect to recruit its strength for a fresh and more violent series of efforts, and to admit of its parts becoming hardened and dry. Suddenly, as we are, perhaps, carelessly looking on the inanimate object, it performs a feat of the most surprising, because unexpected agility. Its body, which was previously much bent backwards, is now swung forwards and bent into a curve; it then swings back, and then forwards again, and so quickly that it almost seems to leap. It then rests its legs upon the front part of the pupa case, and pulls the rest of its body out of the case by degrees, and then creeps forward, leaving the pupa case behind, still immovably fixed upon the plant.