270 [December, 



the wings : but, in all cases, emphatically regard the instructions 

 not to put them within reach of each other's jaws ; even specimens 

 thought to be dead may revive, and then, upon arriving home, the 

 results of a day's labour present no better appearance than jagged 

 wings and mutilated bodies. Though, apparently, so insensible to 

 pain, and tenacious of life, they are, in reality, easily killed, and Dr. 

 Hagen even asserts that they die rapidly by simply placing the box 

 close to a window in the hot sunshine : this I have not tried. 



I now come to the most important joarts of the subject : firstly, 

 the preservation of them in collections without breakage ; secondly, 

 the preservation of the colours. On these points much has been 

 ■RTitten ; and I will endeavour to combine the advice of previous 

 writers with my own experience. The body of a dry dragon-fly is 

 extremely liable to dislocation at any segmental division, the compound 

 consolidated thorax being in reality the only part that will remain en- 

 tire with any degree of certainty. The head is attached very slightly, 

 yet this will often move round, almost as if on a pivot, withoiit be- 

 coming detached : however, if it come off, it can easily be re-attached 

 by a strong solution of shell-lac in spirits of wine ; and some may even 

 prefer to voluntarily behead all their specimens, and unite again with 

 this or some similar liquid cement. Taking the smaller species {Agrio- 

 iiidce, &c.) first, I do not recommend, with them, any system of dis- 

 embowelling, believing, from experience, that the preservation of the 

 colours is not furthered thereby. Dr. Hagen (' Intelligencer,' vol. ii, 

 p. S2, and vol. iv, p. 88) advises that a needle with thread be intro- 

 duced into the under-side of the thorax, and brought out before the 

 termination of the body (mark especially, before the termination, so 

 as not to injure the appendages of the apex), the thread being drawn 

 backwards and forwards to remove some portion of the viscera, and a 

 clean thread then inserted and cut off at each end. Baron de Selys- 

 Longehamps, whose experience is probably greater than that of any 

 other living entomologist in these matters, prefers (' Revue des Odo- 

 nates,' p. 378) to use no endeavours to extract the viscera in the 

 slender species. In this I agree with him, and explain here the plan 

 I adopt. Having provided myself with some stout horse-hairs, I cut 

 them up into short lengths ; one of these lengths is inserted in the 

 under portion of the thorax of the insect, and is gently pushed down 

 until it reaches the extremity of the abdomen, but it is not pushed 

 through, and, by this means, the anal organs can suffer no injury ; the 

 thoracic end is then cut off, and the specimen is thus rendered incapa- 

 ble of breakage. I conceive horsehair, when it can be obtained, to be 



