74 NATURAL HISTORY. 



many minor modifications, the normal structure of the order, may then constitute a great group under 

 the name of DIPTERA GENUINA, which is already sufficiently characterised. 



The classification of the Diptera is a matter of some difficulty, and several different systems have 

 been proposed. Until of late years, however, the order was usually divided into two main sections, 

 the Nemocera and the Brachycera, or long and short-horned types; but other arrangements have 

 been proposed, generally founded, more or less, upon the character of the metamorphosis, the most 

 complete of which is that invented by MM. Brauer and Schiner, which rests almost wholly upon the 

 study of the preparatory states of the insects. This classification, however, we shall not adopt 

 here, partly because the groups thus formed cannot always readily be characterised by peculiarities 

 of the perfect insects, and partly because it is quite uncertain how far minor peculiarities of the 

 larvae and pupae are to be regarded as of importance in classification. The arrangement of the 

 families here followed is very nearly that adopted by Dr. Gerstacker in the " Handbuch der Zoologie," 

 but we have introduced some higher divisions, which are the same as those proposed many years ago 

 by Latreille and adopted by Professor Westwood in his admirable " Introduction to the Modern 

 Classification of Insects." The characters distinguishing the Pupipara, the lowest of our divisions or 

 tribes, from the rest of the Diptera, have already been indicated, and we have therefore now only 

 to consider the " Diptera genuina." These may form four tribes. 



1. NEMOCERA, having the antennae usually composed of from ten to seventeen joints, but 

 sometimes of only six, thread-like or beaded, and the palpi of four or five joints, the cross-veins in 

 the wings usually few or altogether wanting, and the halteres uncovered. The larva is furnished 

 with a more or less distinct head, and gives origin to a free pupa. 



2. NOTACANTHA, having the antennae composed of riot more than twelve joints, which, when 

 most numerous, are sometimes nearly equal, forming a thread-like antenna, as in the preceding 

 division, but more commonly constitute apparently a three-jointed organ, of which the third joint is 

 ringed ; the palpi of not more than three joints ; the veins of the wings usually forked and with 

 cross-veins, and the halteres uncovered. The larva has a more or less distinct head, and the pupa is 

 enclosed in the dried larva skin, from which it escapes, when mature, through a slit in the back. 



3. TANYSTOMA, in which the antennae consist apparently of three joints, but often with 

 indications of articulations in the third joint, and with a terminal bristle ; the palpi of not more 

 than two joints ; and the mouth usually more perfect than in the preceding group. The larvae 

 have a more or less distinct head, and produce free pupae. 



4. ATHERICERA, having the antennae composed of three joints, of which the third shows no 

 indication of rings, but is furnished with a bristle which is often more or less jointed ; the proboscis 

 kneed, usually retracted and concealed when not in use ; the palpi of a single joint, and the 

 mouth with only two or three bristles. The larvae have no distinct head, and the pupa is enclosed 

 within the dried larva-skin, which contracts into an oval form. 



5. PUPIPARA, already characterised (p. 73). 



TRIBE I. NEMOCERA. 



As already described, the Nemocera have thread-like or beaded antennae, and these organs 

 are usually of considerable length, and composed of somewhat numerous joints. In many cases, 

 especially in the males, they are fringed or surrounded with long hairs, which give them a plumose 

 appearance. When the antennae are short, the palpi furnish distinctive chai*acteristics, they being 

 either four- or five-jointed, whereas the palpi of the succeeding tribes never contain more than three 

 joints, one of which is so much more developed than the others that the organs appear one-jointed. 



The insects belonging to this group are numerous and present a considerable variety both 

 of character and habits, and their classification has been variously treated by different authors 

 some regarding them as constituting a single great family, whilst others divide them into from two 

 to a dozen such groups. We shall here adopt a division into seven familio; 



FAMILY I.-CULICID^E, OR GNATS. 



The Gnats or Mosquitoes may be at once distinguished from all the other Nemocera by their 

 long, slender, horny proboscis, which may be half as long as the body of the insect, and is usually 



