118 MODERN CLASSIFICATION OF INSECTS. 



where, in fact, ordinary insect motions, swimming or walking, would 

 be completely paralysed. I have found some of the species in such 

 situations in places where the water is constantly a yard deep, and ob- 

 served that they selected stones with flat rather than uneven sur- 

 faces, and especially those covered with a slight coating of vegetable 

 matter. In company with these insects I discovered a great number 

 of small flat larvae, an outline of one of which I have given at jig. 7. 

 16. upper, 17. underside. The habits of thgse larvae, and their slow 

 movements, induce me to believe them to be the Elmis seneus, which 



A 



I found most plentiful at the same time. I am indebted to W. Raddon, 

 Esq., for a piece of soft whitish mortar -like stone, the under surface 

 of which is very rugose, and burrowed in various directions, in the 

 crevices of which a number of specimens of several species of Elmis 

 were still to be observed ; some of the burrows were formed into re- 

 gular oval chambers, very smooth within, and in some of them I found 

 several rather long and cylindric larvae, with the segments constricted, 

 and with six rather long legs. These larvae I should certainly have 

 regarded as those of the Elmis, had I not discovered those above 

 mentioned, which are more analogous to the perfect Elmis, and had I 

 not also in some of the burrows discovered some of these last-described 

 larvae, each in a fine tube of sand, which induces me to suppose that 

 they must be the larvae of some Neuropterous insect. I have, how- 

 ever, represented one of these larvae \njig. 7. 18. The under surface 

 of the presternum, in the perfect insect, is advanced beneath the 

 mouth {Jiff' 7. 14.), as in the Parnides, so as completely to protect 

 the latter at the will of the insect. We are indebted to M. Leon 

 Dufour for a very valuable memoir upon the external and internal 

 anatomy of Macronychus and Elmis, published in the Annales des 

 Sciences Naturelles, new series, vol. iii., March, 1835. In this me- 

 moir the maxillae of the subgenus Stenelmis are described, and repre- 

 sented as armed with several strong and acute teeth; but from dis- 

 sections which I have made of Elmis aeneus, I am inclined to think 

 that the maxillae are unarmed, and that the appearance of teeth is 

 produced by the transverse matting of the hairs at the tip of the 

 maxillae, as represented in my figure 7. 12. 



Messrs. Dufour and Curtis represent the maxillary palpi as 4- 

 jointed; but under the highest-powered lens and in the strongest light 

 I can only detect three joints. 



The curious little insect composing the G. Georyssus has been 



