Apallaga biseriata maculata ( Hampson, 1891 )

Collins, Steve C., 2017, Observations on the biology of Afrotropical Hesperiidae (Lepidoptera). Part 12. New information and corrections, Zootaxa 4312 (3), pp. 471-496 : 487

Apallaga biseriata maculata ( Hampson, 1891)

As we anticipated ( Cock & Congdon 2011b), A. biseriata (Butler) is a valid species rather than a subspecies of A. galenus (Fabricius) . The population from eastern Kenya which we treated as C. galenus biseriata ( Cock & Congdon 2011b, pp. 28–32, Figures 27–30) is now treated as A. biseriata maculata (Hampson) , which was described from the Sabaki River, eastern Kenya ( Hampson 1891), and occurs from southern Kenya to Zimbabwe ( Libert 2014). Celaenorrhinus handmani Collins & Congdon which we treated as a valid species ( Cock & Congdon 2011b, pp. 32–34, Figures 31–33) is considered a synonym of A. biseriata maculata ( Libert 2014) . We collected both populations from Hypoestes forskaolii (Acanthaceae) and saw no diagnostic differences between the early stages. However, when the barcodes for this material are examined, it is not so clear-cut, as individuals named as A. biseriata maculata can be found within several different barcode clusters, including BINs BOLD:ABY8703, BOLD:AAQ3589, BOLD:ACE5675, BOLD:ACE3207 and BOLD:ABY8700 (BINs are hereafter referred to by the last seven characters), mostly in isolation, but in one case alongside individuals of other species with identical barcodes. We note in particular that specimens from Mufindi, Tanzania, the type locality of C. handmani are placed in a separate BIN (ACE3207) together with two specimens from Ethiopia which appear to be a different species with the same barcode. The bulk of Tanzanian specimens form a separate compact cluster (ABY8703 from Kimboza, Kihansi, Rondo Plateau, Rubeho, Rondo and the Shimba Hills, Kenya). TCEC’s observations on A. handmani and A. biseriatus maculatus indicate allopatric populations at different elevations. Apallaga biseriatus maculatus occurs at lower elevations and seems to occur no higher than at 1800m in the Chyulu Hills ( Van Someren 1939). Apallaga handmani is a moderate to high elevation species, so their ranges do not overlap. Apallaga handmani was thought to occur as low as 1450m on Mt. Mabu in Mozambique (at that latitude, the elevation is equivalent to 1800m or more at the equator), but this population comes out as a different BIN (ACE5675) so there may be additional allopatric diversity here. Again, more work is needed to understand this situation and reconcile taxonomy with barcode patterns (cf. A. opalinus and A. kakamegae above). However, we think the barcode difference justifies reinstating C. handmani stat. rev. as a good species, although not all the type series of A. handmani are likely to represent this species rather than previously unrecognised, very similar but genetically distinct, allopatric species, such as the population from Mozambique mentioned above.