Phyllosphingia dissimilis (Bremer, 1861)

Transferred to Phyllosphingia by Hampson, 1898, J. Bombay nat. Hist. Soc. 11: 280. Zolotuhin & Ryabov (2011) (STI 20664), confirming the presence of Phyllosphingia dissimilis in north and central Vietnam, noted that these moths were larger, more contrastingly patterned and had narrower forewings than specimens attributed to the nominotypical subspecies. Phyllosphingia dissimilis perundulans from NE India was said to differ from the Vietnamese moths in being brighter and with more serrate outer margins to both wings. Furthermore, in Phyllosphingia dissimilis perundulans, the forewing outer margin was straight rather than concave between veins Rs4 and M3. Zolotuhin & Ryabov (2011) also stated that in comparison to the genitalia of a specimen of Phyllosphingia dissimilis perundulans from SE Tibet illustrated by Eitschberger (2002: pl. 30-32; mislabelled as Phyllosphingia dissimilis hoenei) (STI 17730), the harpes of the Vietnamese moths were more robust and symmetrical, differences in other genital structures showing no diagnostically significant differences. No explicit comparison was made with the genitalia of the nominotypical subspecies. Shortly after, in a book on the hawkmoths of Vietnam, Zolotuhin & Ryabov (2012: 97) (STI 20739) reiterated these diagnostic differences but now included the Thailand populations in the same grouping as those from Vietnam. These were now stated to differ from Phyllosphingia dissimilis perundulans in the shape of the forewing outer margin and from the nominotypical subspecies in the symmetrical and more robust harpes. However, in the “Taxonomic Appendix” of the same work, Zolotuhin & Ryabov (2012: 201-202) decided that these differences were, in fact, sufficient to warrant describing the Vietnamese moths as a separate subspecies, Phyllosphingia dissimilis berdievi. No mention was made in the description of whether or not the Thailand populations belonged to the new subspecies. Finally, Zolotuhin & Ryabov (2012: 201) commented that Phyllosphingia dissimilis berdievi might warrant full species status, as the vicariant sister species of Phyllosphingia dissimilis, but refrained from doing so in the absence of evidence from DNA sequence data. However, the habitus differences described by Zolotuhin & Ryabov (2011, 2012) largely disappear when a longer series of individuals is examined. Although moths from NE India lack an indentation on the forewing outer margin between veins Rs4 and M1, there are many specimens in the NHMUK from central China that have a similar forewing shape (as in the moth from “Mien-shan” in MAKB illustrated here). Likewise, although the forewings of the holotype of Phyllosphingia dissimilis perundulans appear rather elongate, this feature is also inconsistent in specimens from the southwest of the species range. Overall, the size, colour and wing shape of Phyllosphingia dissimilis are all exceptionally variable, with small to large, grey to brown and scalloped to even wing edges being found throughout the range of the species and often flying together. As noted by Mell (1935) (STI 19178), colour intensity is dependent upon pupal duration, and size is dependent upon larval duration. It was for these reasons that Kitching & Cadiou (2000) (STI 18788) synonymized Phyllosphingia dissimilis hoenei, Phyllosphingia dissimilis jordani and Phyllosphingia dissimilis sinensis with the nominotypical subspecies. It now seems that Phyllosphingia dissimilis perundulans can be included within this habitus variation, with differences in none of these traits providing any justification for subdividing the species. The symmetry of the harpes in Vietnamese Phyllosphingia dissimilis has not been independently verified due to lack of available specimens and it remains to be determined whether it holds for all populations. Thus, for the present, subspecific status of Phyllosphingia dissimilis berdievi is retained. However, examination of the genitalia of a moth in NHMUK from Thailand (Doi Inthanon) showed the harpes to be asymmetrical. Thus, if the symmetry of these structures in the Vietnamese populations holds true, then the Thailand populations cannot be assigned to Phyllosphingia dissimilis berdievi. They are thus retained here under Phyllosphingia dissimilis perundulans. Zolotuhin & Ryabov (2012) stated that they had no molecular evidence available to them to bring to bear on this issue, which is why they chose to describe of Phyllosphingia dissimilis berdievi at subspecific rank. There are still no DNA barcode data in BOLD (as of 01/11/2016) that inform on the Vietnamese populations, but there are sequences for other populations. There is very little divergence (0.00 – 0.83%, average = 0.45%) between samples of Phyllosphingia dissimilis dissimilis from localities as distant as Honshu, Primorskiy Kray, Taiwan, Shaanxi, Sichuan and Xizang/Tibet. In contrast, a single sample of Phyllosphingia dissimilis perundulans from Doi Inthanon, Thailand, is 3.78 – 4.37% divergent from the more northeastern samples. Although, this level of divergence is widely considered to be indicative of species-level taxonomic separation, it seems not to be unusual within species of Smerinthini. Indeed, maximum intraspecific divergences within species of this tribe can be even higher, e.g., Pseudoclanis postica (3.96%), Smerinthus szechuanus (5.15%), Laothoe populi (5.65%) and Smerinthus kindermannii (6.59%), where the higher-end divergences are associated only with subspecific differentiation at most. Consequently, in the absence of any consistent morphological differences, there is no justification for treating Phyllosphingia dissimilis perundulans as a species and it is retained as a subspecies of Phyllosphingia dissimilis.