In the stem-species pattern of the Chloropidae family-group there is one katepisternal seta. Within the Milichiidae, two setae in the stem-species of Aldrichiomyza+Xenophyllomyza and two or three setae in Eusiphona, in the Milichia speciosa-group, in Pholeomyia, and in Ulia have evolved convergently. It is possible that two or three katepisternal setae are an apomorphy for stem-species H (Milichiinae except Enigmilichia). However, if that is the case then the number of katepisternal setae must have been secondarily reduced to one several times: (1) In Eusiphona, there are three katepisternal setae in Eusiphona sp. 1, but in the other Eusiphona species there is only one katepisternal seta. The presence of three setae is probably the plesiomorphic character state for Eusiphona, because at least in the length of the proboscis Eusiphona sp. 1 shows a less derived character state than the other Eusiphona species, which have an extremely long proboscis. Eusiphona sp. 1 could therefore be the sister-group of the other Eusiphona species. (2) The presence of just one katepisternal seta in Milichiella lacteipennis is probably derived, because there are more katepisternal setae in all other Milichiella species. (3) The presence of one seta in Milichia distinctipennis could be derived, too, because in the M. distinctipennis-group, which comprises four species, there is one species with two katepisternal setae. Since the number of katepisternal setae varies within the Milichiinae, the point of origin of more than one seta can only be determined after the relationships within the subfamily have been resolved by means of other characters (ex Brake 2000).
Diptera (7)
