69 



Pal?earctic and Nearctic regions only. Their characteristics are the 

 long neck in pupa and imago, the habit in the larva of wriggling 

 backwards when alarmed, and the long ovipositor of the female, 

 which is folded along the back in the pupal stage. I have found 

 by experience that the larva of a British species of Raphidia will 

 live for a long time without food, though it shrivels a great deal. 

 The larvai are carnivorous, and appear to feed on insects that live in 

 decaying timber, where the larvae themselves find their home. Just 

 before disclosing the imago, the pupa, which is without a cocoon, 

 runs about, in this respect differing from the pupa of Sialis, which 

 becomes an imago within its cell. 



All the British examples of the family belong to the genus 

 Raphidia. They are four in number. The largest — R. notata — is 

 fairly common in suitable places, and so apparently is R. maculi- 

 collis (PI. V, figs. 2, 2a). R. xanthostigma seems to be scarce, and 

 R. cognata very scarce. The last three are in appearance very similar 

 to one another. Though there is no cocoon which needs opening, 

 the pupa of Raphidia possesses a pair of powerful jaws, which 

 clearly might have been used for that purpose, as, indeed, is the case 

 with the next family. It appears as though these insects have given 

 up making a cocoon, while still retaining the means of opening it — 

 the jaws being, therefore, vestigial. 



Now follow seven families which agree in one respect at least. 

 They all possess a suctorial mouth in the larva, while it is of the 

 biting type in the imago. The larval type of mouth is due to a 

 special modification of the mandibles, whereas, in the Lepidoptera, 

 for instance, in which order the sucking mouth is found in the 

 imago, the modification is in another direction. 



3. MVRMELEONID/E. 



This family is entirely unrepresented in Britain, although one 

 species extends as far north as Southern Sweden. iMyriiteieon 

 formicarius (PI. V, fig. 3). which is found near Paris, is the species 

 with whose life-history Reaumur has made us so well acquainted, 

 especially as regards the pitfall in the form of an inverted cone in the 

 sand, which the larvfe prepare for the capture of small living animals 

 on which they prey. Ants were supposed to be most often captured I 

 presume, whence the whole family are known as ant-lions. A couple 

 kept in captivity for some months, however, would not touch ants, 

 nor did they make satisfactory pitfalls ; but perhaps I did not pro- 

 vide suitable conditions under which they might work. When not 

 fastmg during the winter months they seemed to prefer small cater- 

 pillars for food. One of these made a spherical cocoon of sand and 

 liquid silk fiom which the pupa partly emerged before producing the 

 imago. More than 300 species are known, amongst which there is 

 much variation in size and appearance. Apparently it is only 



