Materials and methods

Specimens examined for this study are from the following institutions / private collections:

BMNH Natural History Museum, London, U.K.;

JA Dr. Joachim Adolphi, Private collection, Dresden, Germany;

RMNH National Museum of Natural History in Leiden, The Netherlands;

RP Robert Perger, Private collection, Santa Cruz, Bolivia;

RV Robert Vigneault, Private collection, Quebec, Canada;

SNSDSenckenberg Naturhistorische Sammlungen Dresden, Germany;

TN Tatsuya Niisato, Private collection, Tokyo, Japan;

UN Ulf Nylander, Private collection, Valbo, Sweden.

Morphological characters were examined with a stereomicroscope and specimens were photographed with a Canon 450D reflex camera fitted with macro lenses.

The following specimens were examined:

Parandrocephalus eversor Heller, 1916

Sumatra, Padangsche Bovenlanden: holotype 1 ♂, RMNH, ex coll. Dr. H. J. Veth; Malaysia, Cameron Highlands, Kampong Rajah: 1 ♂, RP, VI-2000; Borneo, Sabah, Mt. Trus Madi: 1 ♀, UN, 19-VI-05, S.Chew coll.

Parandrocephalus drescheri Blair, 1938

Java, G. Tangkoeban Prahoe, 4000-5000 voet [= ft.]: holotype, 1♂, BMNH, VI-1937, F.C. Drescher coll.[G. Tangkoeban Prahoe, 4000-5000 voet [= ft.], Dreanger, Java, VI.1937 / type / Parandroceph. drescheri Blr., ♂ type det. K.G. Blair / Brit.Mus., 1937-662]

Parandrocephalus blairi Bentanachs & Vives, 2009

Indonesia, Sulawesi, Puncak near Palolo: holotype, 1 ♂, TN, II-1990; Indonesia, Sulawesi, Palolo Palu: allotype, 1 ♀, TN, IV-1991; Indonesia, Central Sulawesi, Palolo Palu: 2 ♂ RP, 1 ♂ JA, III-1999; Indonesia, Central Sulawesi, Puncak: 1 ♂, RP, IV-1999.

Hexamitodera semivelutina Heller, 1896

Indonesia, North Sulawesi: holotype, 1 ♀, SNSD [semivelutina Hh. / N.Celebes, 9389 / Typus / Staatl. Museum für Tierkunde, Dresden]; Indonesia, Sulawesi, Minado: 1 ♂, BMNH, 1781, Wallace coll., ex coll. Fry; Indonesia, Sulawesi: 1 ♀, BMNH, 1922; Indonesia, Sulawesi, Pulu Pulu: 1♂, RV, 19-XII-1997, A. Audureau coll.

Taxonomy

Contrary to the generic characters provided for Parandrocephalus (Bentanachs and Vives 2009), the mandible of Hexamitodera (Sulcognatha) blairi does not possess a dorsomedian carina (Fig. 2A; 3D) and the elytra have prominently (and not weakly) developed carina (Fig. 1 C, D). Consequently the generic position of Hexamitodera (Sulcognatha) blairi had to be reconsidered.

The separate treatment of sexual and non-sexual traits has revealed a suite of apomorphic characters (Table 1) that allows a more coherent interpretation of relationships of Hexamitodera (Sulcognatha) blairi. Actually, the chromatic gender dimorphism (brownish tones in male, blue in female), body pubescence, prominent developed elytral costae, short legs and antennae of Hexamitodera (Sulcognatha) blairi clearly belong to Hexamitodera (Table 1). The relationship becomes already evident in the comparison of the dorsal habitus of the females of Hexamitodera (Sulcognatha) blairi and Hexamitodera semivelutina (Fig. 1D, F).

The lacking sexual dimorphism in mandible and head of Hexamitodera semivelutina suggests an ancestral relationship with Hexamitodera (Sulcognatha) blairi, while the enlargement and complex structure of the mandible in Hexamitodera (Sulcognatha) blairi indicate a more derived status. I hypothesise that the grades of differentiation in the mandible in the two Hexamitodera species represent the basal and terminal end of an evolutionary transformation series (Fig. 3A-D) of male adaptations to adjust on the female during the mating. The interior shape of the male mandible in Hexamitodera (Sulcognatha) blairi fits well with the posterior constriction of the female prothorax, while the antero-distal mandibular concavity is perfectly suited to accommodate the female profemora. The large mandibles and their adjustment on the female during the mating are evolutionarily advantageous because they facilitate a successful (possibly prolonged) fertilization and contribution to the gene pool. In this context, the inclusion of Hexamitodera (Sulcognatha) blairi into Hexamitodera provides an interesting hypothesis for testing evolutionary processes that select for large mandibles and co-adaptations.

It might be asked if the modifications in head and mandible of Hexamitodera (Sulcognatha) blairi justify the establishment of a new genus, however, particularly on the basis of morphology it is difficult to assess at which point a phylogenetic distance passes a subgeneric or generic boundary, and additionally there is no operational definition for such boundaries. Mayr (1969) pragmatically defined a genus as “a taxonomic category containing a single species, or a monophyletic group of species, which is separated from other taxa of the same rank (other genera) by a decided gap.”

To my knowledge, Hexamitodera semivelutina and Hexamitodera (Sulcognatha) blairi are from other taxa of the same rank unambiguously distinguished by apomorphic characters (Table 1) and the lack of intermediate mandible and head forms, indeed representing a certain gap, at the very most justifies the institution of a new subgenus.

Convergence in secondary sexual characters

The extraordinarily developed mandibles in males of Parandrocephalus and Hexamitodera (Sulcognatha) blairi, evidently showing distinct features (Table 1; Fig. 2; 3D, F), should be interpreted as convergent adaptations, possibly for mate-guarding and the fitting of the male mandible to the posterior constriction of the female prothorax. While the mandible structure might contain useful taxonomical information, the enlargement alone is not a good indicator for relationships since it has also independently evolved in males of other Callichromatini genera (e.g. Aphrodisium niisatoi Vives & Bentanachs, 2007 and Huedepohliana superba (Aurivillius, 1910)), and also in phylogenetically more distant beetle taxa, such as Prioninae (Cerambycidae), Manticorini (Carabidae) and Lucanidae.

Heller (1916) considered the opened procoxal cavity in Parandrocephalus and the African Callichromatini genus Dictator Thomson, 1878 (as well having enlarged mandibles) as atypical for this tribe and indicator for a closer relationship, however, such combination is also observed in the phylogenetically distant groups already mentioned before and might represent a convergent co-adaptation. The opened procoxal cavities possibly allow a greater deflection of the procoxa and the strong bend of forebody and abdomen while the mandibles and prolegs keep close contact with the female prothorax during the mating. There might be indeed a Gondwanian relationship between African and Asian Callichromatini taxa, however, I think a detailed phylogenetic analysis based on a larger set of morphological or/and biochemical features is needed to present a robust hypothesis of actual relationships.

Cerambycid-Geography and the Wallace line

Sundaland (including Borneo, Sumatra and Java) is assumed to be zoogeographically separated from ‘Wallacea’ (comprising Sulawesi and the Philippines except for Palawan) (Dickerson 1928) by the Wallace line (Huxley 1868), which is supported by distributional patterns e.g. in mammals and birds (Wallace 1859, 1860; Huxley 1868), cicadas (de Boer and Duffels, 1996), butterflies (see Vane-Wright and de Jong 2003 for a review) and hawkmoths (Beck et al. 2006).

The distribution of Cerambycidae taxa has been only sporadically treated in respect to the Wallace line and there is no statistical approach identifying geographical patterns.

Actually there are several closely related Cerambycidae taxa that occur on both sides of the Wallace line, e.g. the subspecies of Xixuthrus microcerus White, 1853 (Prioninae) (Komiya and Lorenc 2006), species of Komiyandra Santos-Silva, Heffern & Matsuda, 2010 (Parandrinae) (Santos-Silva et al. 2010) and Chloridolum Thomson, 1864 (Callichromatini) (Vives, in litt.), a pattern indicating more recent dispersal and colonization events.

The genera Xystrocera Blanchard, 1845 (Xystrocerini) (Martins 1960; Heffern 2005) and Distenia (Disteniinae) Lepeletier & Serville, 1828 (Gressit 1959; Heffern 2005; Santos-Silva and Hovore 2007) occur not only on both sides of the Wallace line, but also on other continents (Africa and South America), suggesting Gondwanian relationships and longer vicariance processes.

However, in both, the South-East Asian taxa that occur on both sides of the Wallace line and pantropical taxa, distributional patterns might also be influenced by transoceanic dispersal of larvae in drifting logs (see Holzapfel and Harrell 1968; Peck 1994).

The Wallace line holds for the tribe Tmesisternini (Lamiinae), which is highly diversified in New Guinea and Sulawesi but nearly absent in Sundaland (Heffern 2005), and the Callichromatini. Only two (Chloridolum Thomson, 1864, and Pachyteria Audinet-Serville, 1833) (Niisato 2001; Vives and Bentanachs 2010) of the 20 Callichromatini genera that are listed for Borneo (Heffern 2005) are found in Sulawesi. While some of the Chloridolum species that occur on both sides of the Wallace line are morphologically very similar (Vives, in litt.), other species reported for Sulawesi differ from con-generics in Sundaland and might belong to another genus (Vives and Bentanachs 2010).

There appear to be clear trends in some Cerambycidae tribes supporting the Wallace line, nevertheless, the examples predating such line call for a proper statistical analysis of morphological or biochemical characters to clarify phylogeographical relationships.

According to the current state of knowledge, the genus Parandrocephalus did not cross the Wallace line, supporting the idea that Parandrocephalus and Hexamitodera are indeed examples for convergent evolutionary processes within two zoogeographically distinct realms.