Ophrynopus, Konow, 1897

Vilhelmsen, Lars, 2003, Phylogeny and classification of the Orussidae (Insecta: Hymenoptera), a basal parasitic wasp taxon, Zoological Journal of the Linnean Society 139 (3), pp. 337-418 : 382-384

Ophrynopus

Orussonia

Orussella

Orussobaius

Leptorussus

Pseudoryssus

Orussus

Pedicrista

Mocsarya

Chalinus

Ophrynon

Argentophrynopus

Kulcania

Guiglia chiliensis

Guiglia coracina

Guiglia sericata

Guiglia bombycinis

Guiglia rubricata

Guiglia schauinslandi

Guiglia rubicunda

Ophrella

Stirocorsia

Ophrynopus margins of the mesoscutellum (77: 2 i); sistergroup of the ‘ophrynopine’ genera (conc3; k = 2–3 ( Fig. 112 ), supported by the presence of a ventral carina on the hindtibia (105: 1 i); sistergroup of ( Orussus + ( Chalinus + Mocsarya )) + Pseudoryssus (k = 1; Fig. 113 ), supported by the forewing vein 1r-Rs being spectral (120: 0 c). None of these alternatives seem well corroborated in comparison with the preferred topology.

Pseudoryssus is usually placed as the sistergroup of Orussus ( Figs 106 , 107 , 109, 110 , 114 , 115 ). The support is rather weak, putative synapomorphies being the mesoscutellar arm having a raised area without a pit anteriorly (79: 2 i), the forewing discal cell being rhomboid (119: 1 c,i,r), and, under some weighting schemes (conc5–6), the forewing vein 1r-Rs being spectral (120: 0 i). Pseudoryssus rarely comes out as the sistergroup of Leptorussus (conc1–2; Fig. 111 , see also above), but occasionally as sistergroup to ( Chalinus + Mocsarya ) + Orussus (conc3; k = 1–3 ( Figs 112 , 113 ). The latter topology is supported by the same characters as the preferred one, the differences being caused by the conflict between the characters supporting ( Chalinus + Mocsarya ) + Orussus (see above) and (( Chalinus + Mocsarya ) + Pedicrista ) + ‘ophrynopine’ genera.

The ‘ophrynopine’ genera ( Argentophrynopus , Guiglia , Kulcania , Ophrella , Ophrynon , Ophrynopus and Stirocorsia ) constitute a clade that is retrieved under all weighting conditions ( Fig. 108 ). The most convincing synapomorphies are the distal position of the forewing vein 2r relative to the pterostigma (115: 0 i), the presence of a longitudinal carina laterally on the female 9th abdominal tergum (151: 1 i), and the presence of tubercles on the male 9th abdominal sternum (156: 1 c,i). The sistergroup of the ‘ophrynopine’ genera usually is ( Chalinus + Mocsarya ) + Pedicrista ( Figs 106 , 107 , 109, 110 , 114 , 115 ); this topology is rather weakly supported, the strongest support coming from having a carina developed ventrally on the hindtibia (105: 1 i,r) and having a transverse swelling absent from the male 9th abdominal sternum (155: 0 i,r). Rarely occurring alternative sistergroups to the ‘ophrynopine’ genera are all other genera except Orussonia , Orussella and Orussobaius (k = 1; Fig. 113 ), being supported by the characters mentioned above for the clade comprising all genera of Orussidae except the three latter, and (( Leptorussus + Pseudoryssus ) + Orussobaius ) + Pedicrista (conc1–2; Fig. 111 ), a relationship supported mainly by the presence of a carina ventrally on the hindtibia (105: 1 i,r). The support for Pedicrista alone as the sistergroup of the ‘ophrynopine’ genera (conc3; k = 2–3 ( Fig. 112 ) has already been discussed.

The relationships between the ‘ophrynopine’ genera are weakly supported and changes substantially depend on which weighting scheme is implemented. Guiglia is placed as sistergroup of the remaining genera under many weighting conditions, including the preferred topology (conc = 4; k = 4–12 ( Figs 107 , 114 ). The ‘ophrynopine’ genera excluding Guiglia are supported by having the 9th female antennomere swollen subapically (38: 0 i) and a carina present laterally (39: 1 i). An alternative topology (CI; RC; RI; conc1–3, 5–6; k = 1–3 ( Figs 109–113 ) has Kulcania as the basalmost ‘ophrynopine’ genus; the other genera are supported by the presence of a distinct triangular projection on the posteroventral corner of the hindfemur (99: 1). A third possibility is ( Argentophrynopus + Kulcania ) + Ophrynon (k = 13 +; Fig. 115 ) as the sistergroup of the remaining ‘ophrynopine’ genera; the former clade is supported by the antennal characters listed above as support for the ‘ophrynopine’ genera exclusive of Guiglia , the latter clade by the absence of a distinct carina posteriorly on the mesoscutellar arm (80: 0 r).

Argentophrynopus is the most recently described genus in the Orussidae ( Vilhelmsen & Smith, 2002) . A survey of its changing position within the ‘ophrynopine’ clade under different weighting conditions corroborates that it should not be merged with any of the other genera. Argentophrynopus turns up as the sistergroup of (( Ophrynopus + Stirocorsia ) + Ophrella ) + Guiglia rubicunda (RI; Fig. 110 ), of ( Ophrynopus + Stirocorsia ) + Ophrella (CI, RC; conc3, 5–6; k = 3, 6–12 ( Figs 107 , 109 , 112 ), of Ophrynon (conc1–2; k = 1–2 ( Figs 111 , 113 ), and of Kulcania (conc4; k = 4–5, 13 + ( Figs 114 , 115 ). The topology (( Ophrynopus + Stirocorsia ) + Ophrella ) + Argentophrynopus occurs in the preferred tree ( Fig. 107 ) and is supported by the presence of an indistinct lateral margin on the mesoscutellum (77: 0 i); the sistergroup relationship between Argentophrynopus and Ophrynon is supported by the presence of a carina ventrally on the forefemur (56: 1 i). Additional support for these and for the other possible placements of Argentophrynopus is weak, being provided by characters that are higly variable even within the ‘ophrynopine’ clade.

Guiglia is the only genus in the Orussidae that is not retrieved under all weighting conditions. In most cases (CI; RC; conc1–6; k = 1–12 ( Figs 107 , 109 , 111–114 ) the monophyly of the genus is corroborated, although the internal relationships vary; however, Guiglia occasionally comes out as paraphyletic relative to ( Ophrynopus + Stirocorsia ) + Ophrella (k = 13 +; Fig. 115 ) or rarely as polyphyletic (RI), displaying the topology (((( Ophrynopus + Stirocorsia ) + Ophrella ) + G. rubicunda ) + Argentophrynopus ) + remaining Guiglia spp. ( Fig. 110 ). Character support for the monophyly of Guiglia is weak, but so are the characters that corroborate the para- and polyphyletic topologies. I consider it most prudent to retain Guiglia as a valid genus until more evidence can be acquired. The problems with retrieving Guiglia as monophyletic may be caused by missing data for G. rubicunda . This species is only known from a single male specimen, and thus it could not be scored for the characters specific for the female sex. The examination of female G. rubicunda specimens might provide additional support for the inclusion of this species in Guiglia and perhaps strengthen the case for the monophyly of the genus.

The clade ( Ophrynopus + Stirocorsia ) + Ophrella is retrieved under most weighting conditions ( Figs 106 , 107 , 109, 110 , 112 , 114 , 115 ); it is supported by the unique trait of having the mesoscutellum rounded posteriorly and lying parallel with the anterior margin of the metanotum (75: 0). The sistergroup relationship between Ophrynopus and Stirocorsia also comes out in most of the analyses; this is supported by the dorsally extended ocellar corona (6: 0 i) and by the presence of a well developed subspiracular carina (138: 1 i). An additional putative synapomorphy for these two genera is the presence of an occipital concavity dorsally on the head (28: 1 i); however, this trait is absent from Ophrynopus carinatus . In a few instances, Stirocorsia is the sistergroup of Guiglia (conc1–2; k = 1–2), in which case Ophrynopus is either most closely related to the former two genera combined (conc1; Fig. 111 ) or to Ophrella (conc2; k = 1–2 ( Fig. 113 ). The Ophrynopus + Ophrella clade is corroborated by the presence of a lateral metanotal carina (90: 1 i), but otherwise support for these relationships is weak.

Figures

Figure 106. Strict consensus of 540 trees, produced by equal weights, amb- analysis in PAUP*. Numbers at branches are Bremer support values.

Figure 107. Preferred topology, used for mapping character evolution. One of two trees derived by implied weights analysis in PAUP*, k = 7–9, amb-. The tribal classification of Benson (1955a)/Guiglia (1965) is superimposed on the right. Taxa formerly placed in Heliorussus are indicated with asterisks.

Figure 108. Strict consensus tree computed from all trees derived under every weighting condition implemented in PAUP* and Pee-Wee.

Figure 115. Topology of genera of Orussidae, from PAUP* with settings k = 13–18. Note that Guiglia is paraphyletic.

Figures 111–112. Topology of genera of Orussidae, from Pee-Wee with concavity constant 111. conc1, and 112. conc3.

Figures 113–114. Topology of genera of Orussidae, from PAUP* with settings 113. k = 1, 114. k = 4–5.

Figures 109–110. Topology of genera of Orussidae, from successive weightings in PAUP* with 109.CI and 110.RI.