The host of this species was 
established by the discovery of 
puparia in stems of Lithospermum 
arvense and another unidentified 
species of Boraginaceae, 1 km south of 
Poison, Mont., Lake County, 8. IV. 69 
(D. Witwer). 
Melanagromyza proboscidata Spencer, 
new species 
(Figs. 21-23) 
Head (fig. 21). Frons 1.5 times width 
of eye, not projecting above eye in 
profile; 2 ors and 2 ori, latter 
widely spaced, distance between them 
double that between lower ors and 
upper ori, orbital setulae minute, in 
2 rows, those nearest eye margin 
largely upright, inner row more 
proclinate; ocellar triangle narrow, 
apex extending to level of upper ori; 
gena unusually narrow, 0.08 height of 
eye in center, broader in front; 3d 
antennal segment small, rounded, 
arista appearing bare; proboscis 
conspicuously long and slender, 
labellum projecting upward to level of 
antennae in rest position (fig. 21). 
Mesonotum. 2 dc , acr numerous, in 
some 10 rows. 
Wing. Length 2.3 mm, C extending 
strongly to M 1+2, inner crossvein 
just beyond midpoint of discal cell, 
last section of M 3+4 about 0.67 of 
penultimate . 
Color. Frons matt black, ocellar 
triangle and orbits distinctly 
shining; mesonotum shining black; 
abdomen blackish green or coppery; 
squama and fringe white. 
Male genitalia. Aedeagus as in 
figures 22, 23; 9th sternite broadly 
fused at apex; sperm pump with narrow 
blade. 
Host/early stages. Unknown. 
Holotype m. , Texas, Kerr Co., 
Kerrville, 5. IV. 59; paratypes 2 m. , 
3. IV, and 4m., 5 f., 1-11. IV. 59; 
Davis Ranch, NW. Blanco Co., 1 f.. 
22. IV. 59; (all J. F. McAlpine) ; San 
Patricio Co., Welder Wildlife Reserve, 
nr. Sinton, 1 f . , 19-23. III. 65 (J. G. 
C.). Holotype and paratypes in CNC, 1 
m. , 2 f . in KAS. 
Remarks. The long proboscis (fig. 21) 
makes this a distinctive species in 
the group with the abdomen green or 
coppery. The male genitalia are 
characteristic of species in this 
group and show no particular 
differentiation associated with that 
of the proboscis. 
Melanagromyza radicicola Steyskal 
(Fig. 128) 
Melanagromyza radicicola Steyskal, 
1981: 40. Holotype male from 
Maryland in USNM. 
The single male on which the 
description of this species was based 
was found dead in a rearing container, 
in which roots and stems, believed to 
be of Urtica sp., had been kept hoping 
to rear M. martini Spencer, 1969a, 
which is common in Canada. 
The aedeagus is clearly of the type 
occurring in three species feeding in 
seed heads of various genera of 
Asteraceae — M. bidentis (fig. 121), M. 
minima (fig. 123), and M. minimoides 
(fig. 125). It appears to be 
indistinguishable from that of M. 
minimoides . The slight pigmentation 
at the end of the ejaculatory duct is 
also clearly present in specimens 
examined from Bethesda, Md., reared 
from Heliopsis helianthoides (fig. 
126), and also from Ouachita State 
Park, Ark., reared from Rudbeckia sp. 
However, this pigmentation is not 
obviously apparent in the type series 
from the Miami, Fla., area ex 
Verbesina sp. (fig. 125). 
The inner margin of the epandrium of 
M. radicicola has an acute angle and 
two long bristles below the surstylus 
(Steyskal, 1981: fig. 7, A). These 
additional bristles are normally 
lacking in M. minimoides , but one 
paratype from Florida ex Verbesina 
246 
