Rosser, Freitas, Huertas, Joron, Mérot, et al., 2018.
The evolution of reproductive isolation via a switch in mimetic wing coloration has become the paradigm for speciation in aposematic Heliconius butterflies. Here, we provide a counterexample to this, by documenting two cryptic species within the taxon formerly considered Heliconius demeter Staudinger, 1897. Amplified fragment length polymorphisms identify two sympatric genotypic clusters in northern Peru, corresponding to subspecies Heliconius demeter ucayalensis H. Holzinger & R. Holzinger, 1975 and Heliconius demeter joroni ssp. nov. These subspecies are reciprocally monophyletic for the mitochondrial genes COI and COII and the nuclear gene Ef1α, and exhibit marked differences in larval morphology and host plant use. COI sequences from 13 of the 15 currently recognized subspecies show that mtDNA differences are reflected across the range of H. demeter, with a deep phylogenetic split between the southern and northern Amazonian races. As such, our data suggest vicariant speciation driven by disruptive selection for larval performance on different host plants. We raise Heliconius demeter eratosignis (Joicey & Talbot, 1925) to Heliconius eratosignis based on nomenclatural priority, a species also comprising H. eratosignis ucayalensis comb. nov. and three other southern Amazonian races. Heliconius demeter joroni spp. nov. remains within H. demeter s.s., along with northern Amazonian and Guianan subspecies.
Keywords: butterflies, cryptic species, genotypic clusters, host plant shift, integrative taxonomy, mimicry, vicariant speciation
Heliconius demeter joroni Lamas & Rosser ssp. nov.
Heliconius demeter [ssp. nov.] Lamas, MS: Lamas, 2004: 268. Lamas & Jiggins, 2017: 224.
Diagnosis: Heliconius demeter joroni ssp. nov. is similar to H. demeter demeter, but differs from Staudinger’s syntypes of H. demeter from Iquitos, Loreto, Peru (now in the MNB) in having a much narrower yellow postmedian band on the dorsal forewing. It is known only from the Cordillera Escalera, near Tarapoto, Peru (Fig. 1), where its co-mimics include Heliconius eratosignis ucayalensis, Heliconius elevatus pseudocupidineus Neustetter, 1931, Heliconius aoede cupidineus Stichel, 1906 and Eueides tales michaeli Zikán, 1937, among others. Males are easily distinguishable from all sympatric taxa through the fused rays on the hindwing dorsum and the yellow costal streak on the forewing underside. Females may be distinguished from co-mimics through the configuration of the rays (which radiate from the cell), small size, length of the antennae (longer than the forewing discal cell) and the presence of the forewing underside yellow costal streak. Both sexes usually exhibit a single row of white submarginal dots along the anal margin of the ventral hindwing, which can be used to help separate the females from H. erato emma and E. tales michaeli. This character can be faint or even missing in H. demeter joroni ssp. nov. and occasionally present in H. erato emma. However, the latter is confined to the Amazonian lowlands adjacent to the Cordillera Escalera, and at present there is no evidence to suggest that they regularly co-occur, barring occasional migrants.
Etymology: The subspecies name (a masculine noun in the genitive case) recognizes the contribution of the French evolutionary biologist Dr Mathieu Joron to the knowledge of the mimetic butterfly fauna of San Martín, Peru. Dr Joron is presently a Senior Scientist at the Centre d’Ecologie Fonctionnelle et Evolutive in Montpellier. He began studying the butterflies of San Martín during his PhD and has continued to do so throughout his career, with a particular focus on Heliconius numata.
Geographic distribution: Subspecies of H. demeter and H. eratosignis are mapped in Fig. 1, with photos of a type specimen of each race. Races of H. demeter occupy the Guianas and much of the Amazon basin. H. eratosignis races occur in the west and south of the Amazon basin. In Tarapoto, the two species fly together at a number of sites in the Cordillera Escalera. Only H. eratosignis has been recorded from the adjacent Amazonian lowlands, despite considerable sampling in the area. Museum data and observations by Keith Brown (1979) suggest that the two overlap (at least broadly) in the extreme south of Pará and northern Mato Grosso, in Brazil. There may well also be a contact zone on the Juruá River, between Porto Walter and Eirunepé, as both H. demeter demeter and H. eratosignis tambopata are known to occur there. However, the exact position of contact in this very large area is unclear. In data published by Brown (1979) two additional contact zones are indicated, at Pucallpa, Peru and near Cobija on the Brazilian/Bolivian border. We were unable to locate the relevant specimens in museum collections; however, we consider these points unreliable and excluded them from the distribution map in Fig. 1. The first is probably a generalized locality, with the specimens potentially coming from a large area of northern Peru. The second is likely explained through the co-occurrence of both H. eratosignis ulysses and H. eratosignis tambopata, as the latter was not described at the time (Lamas, 1985).
Neil Rosser, André V. L. Freitas, Blanca Huertas, Mathieu Joron, Gerardo Lamas Claire Mérot, Fraser Simpson, Keith R. Willmott, James Mallet and Kanchon K. Dasmahapatra. 2018. Cryptic Speciation associated with Geographic and Ecological Divergence in Two Amazonian Heliconius Butterflies. Zoological Journal of the Linnean Society. zly046. DOI: 10.1093/zoolinnean/zly046