Dracontomyia Becker

Norrbom, Allen L., Sutton, Bruce D., Steck, Gary J. & Monzón, José, 2010, New genera, species and host plant records of Nearctic and Neotropical Tephritidae (Diptera) 2398, Zootaxa 2398, pp. 1-65 : 21-22

Dracontomyia Becker

Figs. 37–39

Dracontomyia Becker 1919: 193 (type species D. riveti Becker , by monotypy).

Cecidocharella Hendel 1936: 74 (type species C. elegans Hendel , by monotypy). New synonymy.

Cecidocharella Hendel is here considered a subjective junior synonym of Dracontomyia Becker , and the three species originally described in the former genus are transferred to Dracontomyia . These include D. tucumana ( Aczél 1953: 118) , new combination, from Argentina, D. borrichia ( Bush & Huettel 1970: 89) , new combination, from USA (southern coastal Texas), and D. elegans ( Hendel 1936: 75) , new combination, from southern Brazil, Paraguay, and northern Argentina. The genus also includes D. footei Aczél ( Ecuador, Peru), D. riveti Becker ( Ecuador) , and three undescribed species ( Argentina, Bolivia).

Dracontomyia species breed in Asteraceae . At least two of the South American species are gall-formers (Norrbom, unpubl. data), but D. borrichia breeds in flowers of Borrichia frutescens ( Bush & Huettel, 1970). The described species were keyed by Aczél (1953) and Bush & Huettel (1970, as Cecidocharella ).

Aczél (1953), Bush & Huettel (1970), and Foote (1980) used several wing characters to distinguish Dracontomyia and Cecidocharella , but they all intergrade when the three undescribed species are included. It should be noted that fig. 42 of Foote (1980) of the wing of “ Cecidocharella sp. ” is Stenopa mexicana , and fig. 58 of “ Dracontomyia footei ” is D. riveti . The shape of the pterostigma is strongly different only in D. footei ; it is subtriangular in D. riveti ( Fig. 39) and one new species, intermediate between the triangular shape in D. footei and the subrectangular shape in the remaining species ( Figs. 37–38). Vein R 2+3 length and the related ratio of distance between the apices of R 1 and R 2+3 / distance between the apices of R 2+3 and R 4+5 are intermediate in D. elegans ( Fig. 38), D. tucumana , and two new species (ratio 1.5–2.0), shorter than in D. borrichia (2.30–2.42) ( Fig. 37) but longer than in D. footei , D. riveti ( Fig. 39) and one new species (0.76–1.38). Foote’s (1980) use of the length of “cell R” (= cell br) probably was an error referring to the shape of cell r 1, which is also directly related to the length of vein R 2+3. The bulla at the base of R 2+3 is present only in D. footei and D. riveti ( Fig. 39), and that in cell r 4+5 anterior to crossvein dm-cu varies from weak to absent in D. borrichia , to weak in D. elegans and one new species, moderate in two new species, and strong in D. footei and D. riveti .

In addition to the intergradation of the previously proposed diagnostic characters, D. elegans (the type species of Cecidocharella ) and D. tucumana (the latter not examined, based on Aczél’s description) appear to be more closely related to the other species of Dracontomyia exclusive of D. borrichia . Thus Cecidocharella is paraphyletic if D. borrichia is included. Rather than proposing another genus for the latter species, it is preferable to synonymize Cecidocharella with Dracontomyia . Further study of the relationships of D. borrichia to test if it truly belongs in Dracontomyia would be useful.

Several probable synapomorphies support the closer relationship of D. elegans and D. tucumana to the other species of Dracontomyia than to D. borrichia : ratio of distance between apices of R 1 and R 2+3 / distance between apices of R 2+3 and R 4+5 less than 2.0; cell dm broad, length of dm-cu/ width of radial cells anterior to it at least 0.85 (<0.60 in D. borrichia ); distance from bm-cu to r-m along vein M/ distance from bm-cu to dm-cu less than or equal to 0.5 (0.60–0.66 in D. borrichia ); cell m proximal marginal hyaline mark touching or almost touching dm-cu (well separated in D. borrichia ); and mediotergite mostly to entirely microtrichose (except mostly nonmicrotrichose in one new species). One character conflicts with this hypothesis, instead supporting the closer relationship of D. elegans , D. tucumana and D. borrichia : scutum with postsutural sublateral shiny nonmicrotrichose area (also present in D. tucumana ; and common in Cecidocharini , e.g., Neorhagoletis , many Cecidochares and Procecidochares ).

The following are probably autapomorphies of D. borrichia : anepisternum with shiny nonmicrotrichose area anteriorly and katepisternum with small nonmicrotrichose area ventral to katepisternal seta (one or both of these sclerites are partially nonmicrotrichose in some Cecidocharini , e.g., Ostracocoelia , some Cecidochares and Procecidochares ); vein R 4+5 nonsetulose; hyaline band across middle of wing slightly oblique, at posterior wing margin including apex of vein A 1 +Cu 2 ( Fig. 37); and apical band not separated from costa by hyaline area.

Possible synapomorphies of Dracontomyia are the yellow anepimeral seta (also occurs in Phacelochaeta , n. gen., and some Cecidocharini ) and the similar wing pattern (4–5 bands, apical 3 forming F-shaped mark), although the latter may be a synapomorphy with Stenopa Loew and is not uncommon in other tephritid genera.

Dracontomyia has been included in the tribe Cecidocharini but along with Stenopa was considered not to belong there by Korneyev (1999). Phacelochaeta , n. gen. (based on species previously included in Cecidochares ) also may be related. The lateral vertical seta in these genera is no more than half as long as and often paler brown than the medial vertical seta (in the Cecidocharini it is subequal and the same color as the lateral vertical), and the postocular setae are mixed small, acuminate and larger, lanceolate, synapomorphies of the “Higher Tephritinae ”, which does not include the Cecidocharini ( Korneyev 1999) . Another character that supports their removal from the Cecidocharini is their anteromedially setulose frons (except in D. borrichia ; it is nonsetulose in most Cecidocharini ).

Dracontomyia , Stenopa and Phacelochaeta may be closely related, although it should be noted that the putative synapomorphies for this clade also occur in some Cecidocharini . Synapomorphies for Dracontomyia + Stenopa + Phacelochaeta include: scutum with clusters of yellow lanceolate setulae near posterior margin, including at least 1 pair anterior to corners of scutellum (absent in S. vulnerata (Loew) and poorly differentiated in 1 undescribed species of Dracontomyia ; common in Cecidocharini , e.g., Neorhagoletis with 2 pairs, Hetschkomyia with 1 pair, Cecidochares with 0–2 pairs); scutellum with clusters of lanceolate setulae, at least 1 pair basal to basal seta (absent in 3 species of Dracontomyia and variable in S. vulnerata ; also common in Cecidocharini ); microtrichia dark brown, contrasting with gray to tan color elsewhere, on posterior margin of scutum, scutellum, and/or subscutellum (occurs in some Cecidocharini , e.g., Neorhagoletis ); scutellum strongly convex, shiny nonmicrotrichose medially (sparsely microtrichose in Stenopa ; also occurs in most Cecidocharini ); and wing banded and wing base (cell bc and base of br) infuscated (also occurs in most Cecidocharini ). The wing patterns of Dracontomyia ( Figs. 37–39) and Stenopa ( Fig. 96) are especially similar, with the three most apical bands forming a somewhat F-shaped pattern (as in Neorhagoletis ), whereas these bands, particularly that covering dm-cu, are oblique in Phacelochaeta ( Figs. 54–56).