First non-Australian record of pollen-feeding Phasmodines: a new Tettigoniid from the late Paleocene of Argentina

PETRULEVICIUS J. F.

Riberão Preto

Congreso; I International Meeting on Paleoarthropodology, I Simposio Sudamericano de Paleoartropodología; 2000

Extant Phasmodinae are endemic from Australia. Recent discoveries on the late Paleocene of north-western Argentina could give us the first fossil representative of the taxon. On this way, presence of Phasmodines would go back at least to the Late Cretaceous, when Australia and South America were both connected by Antarctica in Gondwanic times. Phasmodinae seems to be the sister group of Meconematinae (Gorochov, 1988), and their unique discovered synapomorphy [Stridulatory veins merging at base like a horseshoe] is present in one of the studied specimens. Phasmodinae include five genera: Zaprochilus, Anthophiloptera, Windbalea, Kawanaphila and Phasmodes. The main synapomorphy of living species of the taxon is the long and triangular prognathous head (Gorochov, 1995). This character is not preserved in the studied specimen; however the latter shares with the winged species a short ScA ending at a costal lobe (convergently present in the genus Clonia, Saginae), which is considered as a synapomophy. Representatives of Phasmodinae have a unique feeding habit among living Tettigoniidae: Phasmodes species feed on the entire flower and the other genera on pollen and nectar. It is quite possible that the latest genera may be involved in pollination (Rentz, 1993). Phasmodinae species have specialised mouthparts and alimentary tracts adapted for the floral feeding specialization. On other hand, pollinivory on Classopollis has been found in ancestors of Tettigonioidea (Aboilus; Krassilov et al., 1997) in the Jurassic of Kazakhstan, whereas it worth noting the presence of this pollen genus in the environment of the studied specimen (Petrulevièius, 1996). If phylogenetic hypothesis proposed by Gorochov (1995) is correct, pollinivory of the Phasmodinae could be a derived feature. Otherwise, ?non pollinivory? has to be acquired at least three times in the taxon phylogenesis, what is more improbable. If we question whether this fossil species could eat pollen and, of course, have the specialised state characters related, the answer could be given after a phylogenetic hypothesis be done, and the situation of the fossil species respect to living species be determined. If we think in the case that the fossil taxon is the sister group of all living species of Phasmodinae (what could be the more probable), and pollinivory is a derived attribute, the probability for the event is 0.5 and do not allow any prospective for the inference of unknown characters and attributes (Nel, 1997).