210 ADErHAGA. [Gijrinidce. 



of the Phytophaga, and seem to point to a partially vegetable diet ; the 

 lower mouth organs are only slightly of the predaceous type ; the sharp 

 sickle-like maxillse seem certainly predaceous, l)ut they have the outer 

 lobe always undivided, and occasionally (e.g. Dineutes) entirely wanting ; 

 the former peculiarity, however, is found among the Carabidsein CaUistus 

 {vide p. 32). 



2nd. The Gyrinidte possess a pair of eyes on each side of the head, on 

 the upper and lower surface ; this appears to be a development caused by 

 their habits, as they swim on the surface of the water, and so require to 

 be on their guard against enemies both from above and below ; this 

 peculiarity however, according to Dr. Sharp (Ent. Mo. Mag. v. 52), is 

 repeated in one of the Carabidaj, Adelotopus., which is found under bark 

 of trees, and seems to have no need, as far as we can judge, of such an 

 arrangement. 



3rd. The antennre are quite different from those of the Adephaga 

 generally, being very short and thick, with the first joint small, the 

 second large and dilated, the third large and inserted at the side of tlie 

 second, and the rest very much compressed and so soldered together that 

 it is difficult to determine of how many joints they really consist ; in 

 many respects they resemble the antennae of the Parnidse. 



4th. The structure of the legs is entirely different ; both the posterior 

 and middle pairs are modified so as to form swimming paddles, and 

 are arrano-ed so that they can be packed away underneath the surface of 

 the body and offer not the least resistance to rapid motion, while l)y 

 thrusting out the one or the other leg the insect can at once change its 

 course ; hence their rapid motions on the surface of the water ; in the 

 Dytiscidse the hinder pairs only are modified. 



5th. In the members of the Adephagous series generally five ventral 

 seo-ments are visible extending right across the body, a sixth being 

 visible only on each side ; in the Gyrinidae, however (as shown by Dr. 

 Horn, Carabidse, p. 93), the first segment is very long, and is composed of 

 two segments united, the rudiments of the suture being distinctly visible ; 

 they have, therefore, six ventral segments reaching across the body, and a 

 seventh visible only on each side. 



The Gyrinidse bear a strong external resemblance to the Pseudomor- 

 liltina, one of the most aberrant of the Carabida?, but it is merely super- 

 ficial ; the peculiarity of Adeloiopus, Avhich belongs to these, has been 

 before alluded to; they are connected with the Dytiscidse by the entire 

 absence of the ante-coxal piece before the hind coxas, which is large and 

 distinct, and reaches across the greater part of the body in the Carabidee 

 and Haliplidae, is very small in the Amphizoida3 and Felobiidse, and 

 entirely wanting in the Dytiscidse and Gyrinidae. 



Dr. Horn is strongly of opinion that in spite of all that may be urged 

 to the contrary the Gyrinidse must be retained in connection with the 

 Adephaga ; Dr. Sharp (Trans. Ent. Soc. 1882, i. 69) is inclined to hold 

 the contrary opinion ; Tliomson (Skand. Col. i. 113) places them with 

 the Parnidie and Elmidre in his group Amphibii between the Palpicornia 



