8o 
Psyche 
[March 
Wirth et al (1965) have used the larval state to define the first 
phylogenetic divergence and thus to define subfamilies of the syrphids 
(Plan 3). This arrangement too creates more problems than it 
solves. Placing the syrphine before the microdontine divergence leaves 
no synapomorphic characters for the Milesinae and creates even more 
convergences than Plan 2. 
In short, Plan 1 seems to offer the most logical illustration of the 
relationship of the Microdontinae to the other syrphids. However, 
much is still to be learned about the phylogeny of the Syrphidae, and 
my placement of the Microdontinae must be accepted only as the 
best possible present arrangement. The strongly plesiomorphic na- 
ture of the subfamily suggests that the microdons might best be 
considered as a separate family (as Martin (1968) has done with 
the Leptogasteridae) . However, regardless of the phylogenetic posi- 
tion of the microdontine flies, they should be clearly recognized as a 
subfamily equivalent to the Syrphinae and Milesinae. 
No other groups have been derived from the microdontine line. 
Hull (1949) included the Eumerinae 4 and Nausigasterinae 4 in the 
microdontine divergence. However, these groups belong to the mile- 
sine line and are probably derived from a myoleptine ancestor. 
Eumerinae and Nausigasterinae could not have evolved from the 
Microdontinae for a number of reasons. Both of these groups lack 
a number of the specialized characteristics of the Microdontinae 
which one would expect to find in any derived group; for example, 
they lack 1) a complete postmetacoxal bridge, 2) the dorsal infold- 
ing of the penis sheath, 3) the double sustentacular apodeme or its 
absence, and 4) other genitalic characters. It is also highly unlikely 
that the phytophagous larvae of Eumerinae and the saprophytic larvae 
of Nausigasterinae could have evolved from a specialized larval form 
like Microdon which lacks segmentation and segmental spines and 
possesses specialized mouthparts. 
Provisional key to the New World genera and 
SUBGENERA OF MICRODONTINAE 
i. Abdomen petiolate; metasterna undeveloped, reduced to a thin 
line and bare (Fig. 5) Mixogaster Macquart 
Abdomen usually not petiolate; if petiolate then metasterna well- 
developed, not reduced and usually pilose (Fig. 6) 2 
4 The use of these groups as subfamilies follows Hull (1949) ; I presently 
regard these two groups as forming one tribe with Merodon , Alipumilio 
and Psilota under the Milesinae. 
