CLASSIFICATION OF DIPTKRA 19 



produce simple keys, amply illustrated. It must be remembered, how- 

 ever, that the illustrations will not serve for all si)eeies in a genus but 

 represent, as a rule, typical forms. 



No use has been made, in general, of subfamilies or subgenera. 

 The use of such almost always calls for many exceptions and also un- 

 necessarily encumbers the literature. A sub-family is supposedly a 

 distinct entity within a family, a group separable upon certain char- 

 acters, but that such classification is unnecessary is evidenced by the 

 exceptions. As an example — the Tachinidse have been divided into 

 many sub-families, or even families, upon such characters as the absence 

 of abdominal bristles, shape of head, etc., etc., but there is not a single 

 character, nor for that matter group of characters, upon w^hich any of 

 these groupings may be retained. Many of the characters in this family 

 are excellent up to a certain point, but every character falls down, not 

 once, but in many instances throughout the family. It is entirely im- 

 possible to separate the Tachinida? and Dcxiidff, just as it is to separate 

 the Muscidffi, Anthomyida? and Scatophagida?. 



Subgenera are superfluous. They are proposed as a rule because 

 the author hopes that some day some character may be found upon 

 which a distinct genus may be based. It is well to remember that in 

 nature there is no such thing as a subgenus, genus or family. Their 

 use is a human creation made possible because links connecting the 

 groups have either entirely disappeared or have not been discovered 

 by man. Genera are therefore artificial and the answer to the question 

 of ''what is a genus?" is that it is something limited in some way by 

 some one. This may sound facetious but it is not meant to be so. 

 The trutli is that certain arbitrary characters have been set up, all 

 the species falling within the limits set forth constituting a genus (or 

 a species or family). It is presumed, but by no means certain that 

 connecting links do not exist. 



Many people describe sub-species, varieties, races, forms, etc., espe- 

 cially in the Lepidoptera, but, fortunately, the fad has not become ex- 

 tensive in Dipterology and it is to be hoped that it may never do so. 

 In actual fact there is no difference in a sub-species, race or variety. 

 A variety is something which differs from the typical or first described 

 form in some definite way. Thoy are explained as potential species, 

 ignoring the fact that no two specimens are exactly alike and that if two 

 extremes of similar type should mate a new race would be formed. In 

 each batch of eggs there are produced forms possessing characters some- 

 what different from the par(-nts. But the species does not change be- 

 cause the average is maintained by the resultant mating. It is only 

 when forms become more or less isolated that a definite trend occurs and 



