Lopidea robusta (Uhler)

Forero, D., 2008, Revision And Phylogenetic Analysis Of The Hadronema Group (Miridae: Orthotylinae: Orthotylini), With Descriptions Of New Genera And New Species, And Comments On The Neotropical Genus Tupimiris, Bulletin of the American Museum of Natural History 2008 (312), pp. 1-172 : 151-152

Lopidea robusta (Uhler)

Hadronema robusta Uhler, 1894: 250 [n. sp.].

Lopidea robusta: Van Duzee, 1928: 182 [new combination]; Carvalho, 1958: 87 [catalog]; Knight, 1962: 32 [discussion, paramere illustration]; Schuh, 1995: 138 [catalog].

DISCUSSION: Van Duzee (1928), while describing Hadronema uhleri , mentioned that Hadronema robusta Uhler was in fact a Lopidea species. Knight (1962: 32) stated that he examined the ‘‘type’’ of L. robusta in the collection of the California Academy of Science (CAS) and gave an illustration of the right paramere (his fig. 16). In doing this, Knight validated a lectotype designation for L. robusta (ICZN article 74.5.). The examined specimens from CAS have affixed lecto- and allotype labels of ‘‘ robusta ’’, which match Uhler’s description and locality data, and the male matches Knight’s paramere illustration. It is probable that Van Duzee affixed those lecto- and allotype labels while describing A. uhleri (see Van Duzee, 1928). Asquith (1991) did not treat or discuss L. robusta in his monograph of Lopidea , because this species is only known from Baja

California in Mexico, an area outside the scope of his paper.

LECTOTYPE MALE: MEXICO: Baja California Sur: San Jose del Cabo, [23.05 ° N 109.68333 ° W], [G. Eisen], Uhler type, Lectotype ‘‘ robusta ’’, ‘‘ Hadronema robusta Uhler San José del Cabo’’, California Academy of Sciences Type No. 551, 13 (AMNH_PBI 00077897) (CAS).

PARALECTOTYPE: MEXICO: Baja California Sur: San Jose del Cabo, [23.05 ° N 109.68333 ° W], [G. Eisen], Uhler type, Allotype ‘‘ robusta ’’, California Academy of Sciences Type No. 552, 1♀ ( AMNH _ PBI 00077898 ) ( CAS) .

PHYLOGENETIC ANALYSIS

TAXA

For the phylogenetic analysis, 24 terminals were considered: 5 as outgroups and 19 as the ingroup. All the species of Hadronema group were included in the phylogenetic analysis. Only H. cinerescens , a fossil species of uncertain position (see above) ( Scudder, 1890), was excluded. For the outgroup, five species in four genera of the Orthotylus group of Schuh (1974) were selected: Lopidea robiniae ( Uhler, 1861) , Orthotylus marginalis Reuter, 1883 , Slaterocoris atritibialis ( Knight, 1938) , and two unidentified species of Araucanocoris Carvalho, 1983 . The tree was rooted with O. marginalis . Because a phylogenetic framework for the Orthotylini has not yet been proposed, a justification for the selection of the outgroups seems necessary. Lopidea externally resembles Hadronema , in particular the head structure and coloration ( Asquith, 1991). Some affinities have been proposed between Lopidea and Hadronema ( Knight, 1968) , although later rejected, but not under a phylogenetic framework ( Schuh, 1989; Asquith, 1991). Lopidea and Slaterocoris have in turn been proposed as related ( Kelton, 1959). Orthotylus marginalis was included, as it is the type species of Orthotylus Fieber , which is the type genus of the subfamily and nominal tribe. Araucanocoris was included to have a Neotropical representative. No hypotheses of relationship have previously been proposed for the taxa considered here.

DATA

A morphological matrix with 69 characters, coded from the vestiture, head, thorax, legs, and male and female genitalia (tables 3, 4), was constructed using Winclada ( Nixon, 2002) as the interface. Characters were mainly binary ones (79.5%) (table 4). Those coded as multistate (20.5%) were considered as nonadditive, as is common for morphological characters (e.g., Goloboff, 1997). Inapplicable data account for 4% of the total information coded in the matrix. Some characters were uninformative, so they were deactivated from the matrix and not used during the analysis. In the consensus tree (fig. 49), all characters supporting nodes in the unambiguous optimization are presented.

ANALYSIS STRATEGY

Parsimony was used as the optimality criterion. All tree searches were completed in NONA ( Goloboff, 1999) spawned from WinClada ( Nixon, 2002). A heuristic search was carried out holding a maximum of 20,001 trees in memory, with 1000 replications and 20 trees to hold per replication, in random addition sequence. For each replication a Subtree Pruning and Regrafting and a Tree Bisection and Reconnection swapping were performed ( Swofford et al., 1996). In the final trees all unsupported branches were collapsed.