Arsenurinae

De, Amabílio J. A., Mielke, Olaf H. H. & Casagrande, Mirna M., 2009, Cladistic analysis of the subfamily Arsenurinae (Lepidoptera, Saturniidae) based on adult morphology, Zootaxa 2218, pp. 1-34 : 28-29

Relationshipss between Arsenurinae genera

Except for brief proposals by Michener (1952), the tribal separation (Almeidaiini and Arsenurini) presented by Lemaire (1980) and work by Peigler (1993), the relationship among the ten genera within Arsenurinae have been incompletely presented and discussed. The results of this study corroborate Peigler (1993) only partially, with different topologies for some clades, possibly given the characters used and sample universe. The analysis resulted in the following relationship hypothesis among the genera: ( Almeidaia (( Loxolomia , Copiopteryx ) ( Rhescyntis (( Grammopelta , Arsenura ) ( Caio ( Paradaemonia ( Dysdaemonia , Titaea ))))))).

It is perhaps not surprising that the genus Almeidaia forms a separate clade, given the distinct morphological aspects compared to other Arsenurinae . This result is in agreement with Lemaire (1980), who included the group in a separate tribe. According to Lemaire (1980) and Peigler (1993), Almeidaia shares affinities with Ceratocampinae , and could be designated as a separate subfamily, being apparently a primitive group.

The relationship between Loxolomia and Copiopteryx is supported by the common origin of M3, CuA1 and CuA2 (character 21:1). Similar to Peigler (1993), Rhescyntis appears neighboring this clade, although belonging to another large clade formed by the other genera.

Grammopelta and Almeidaia have been considered as the most primitive genera among the Arsenurinae ( Michener 1952; Lemaire 1980), so the less basal position of Grammopelta and its status of sister group of Arsenura was not expected. These two genera present several distinct characters and the support of this clade is only 51% according to Jackknife analysis. However, the basic form of the wings, more evident on males and especially with less accented subapical concavity, had some influence on the final result of the analysis. According to Peigler (1993), the relationship between Grammopelta and Loxolomia is even less supported (17%), and he indicated this is probably a false sister-group.

The relationship between Caio , Dysdaemonia , Titaea and Paradaemonia is supported basically by the tibial spur formula (character 27:1). This result was expected given the general appearance of the adults, especially between Titaea and Dysdaemonia . Except for Paradaemonia , whose larvae feed on Lythraceae , there are records that larvae of the species of these genera are specialized in Bombacaceae (Travassos & D’Almeida 1937; D’Araujo e Silva et al. 1968; Dias 1978; Janzen 1982; Stone 1991; Peigler 1993). The relationship hypothesis among these genera is in agreement with the results presented by Peigler (1993).

The relationship proposed by Michener (1952), where the genera Arsenura , Dysdaemonia , Titaea and Paradaemonia are presented as subgenera of Rhescyntis , and also the association between Loxolomia and Grammopelta proposed by Peigler (1993), were not corroborated by the cladistic analysis.