Bombus rufipes Lepeletier de Saint-Fargeau, [1835] Bombus flavipes Handlirsch, 1888: 225 Bombus rufipes Friese, 1914: 10 Bombus rufipes Bremus rufipes Frison, 1930: 6 Widespread polytypic species or complexes of local species? Revising bumblebees of the subgenus Melanobombus world-wide (Hymenoptera, Apidae, Bombus) Williams, Paul H. Altanchimeg, Dorjsuren Byvaltsev, Alexandr Jonghe, Roland De Jaffar, Saleem Japoshvili, George Kahono, Sih Liang, Huan Mei, Maurizio Monfared, Alireza Nidup, Tshering Raina, Rifat Ren, Zongxin Thanoosing, Chawatat Zhao, Yanhui Orr, Michael C. European Journal of Taxonomy 2020 2020-10-02 719 1 120 MFTW Lepeletier, 1835 Lepeletier de Saint-Fargeau 1835 [600,988,264,291] Insecta Apidae Bombus Animalia Hymenoptera 47 48 Arthropoda species rufipes   Figs 12, 35–37, 182    Bombus rufipesLepeletier de Saint-Fargeau, [1835]: 473.     Bombus flavipes Handlirsch, 1888: 225.    Bombus rufipesvar. [subsp.] obscuripes  Friese, 1914: 10.   Bombus rufipesvar. [subsp.] intermissusFriese, 1918: 516.    Bremus rufipesvar. [subsp.] richardsi  Frison, 1930: 6.    Bombus rufipesand  B. eximiusare now considered parts of the subgenus  Melanobombus(see the comments on  B. eximius). Our PTP analysis ( Fig. 10) of coalescents in the COI gene within the rufipes-group supports two species  B. eximiusand  B. rufipes, corroborated by differences in morphology. From morphology, the black wings, black hair of the male face, and the matching form of the male genital volsella and gonostylus support as conspecific the individuals from Sumatra, which usually (not always) have the mid and hind tibiae orange (taxon obscuripes), together with the individuals from Java, which more usually have the mid and hind tibiae predominantly black (taxon  rufipes s. str.). The gonostylus interior proximal process is less strongly bifid for the male from Sumatra. For the male labelled Borneo, the black wings, black facial hair, and form of the volsella and gonostylus all match  B. rufipes. For this male, the hair of the mid and hind basitarsi is orange, although the hair of the mid tibia is black and the hair of the hind tibia is orange at the base and black at the tips. A few individuals from Sumatra and Peninsular Malaysiahave the hair of the scutellum brown or orange (taxon richardsi). A queen of ‘ostindischen’ origin is described as having T5–6 orange-red (taxon  flavipes, not seen).    Diagnosis  Females  Queens large body length 24–26 mm, workers 14–18 mm. Can be distinguished by their combination of the hair of the metasoma entirely black with the wings nearly black with the veins black (cf.  B. eximius), the hair and integument of mid and hind tibiae and of all barsitarsi either orange or black. The labral lamella has the anterior edge broad and nearly straight (cf.  B.( Megabombus) senexVollenhoven, 1873,  B.( Megabombus) melanopodaCockerell, 1910, from Sumatra).  Males Body length 15–18 mm. Can be distinguished by their combination of the hair of the face black with the metasoma entirely black with the wings nearly black with the veins black (cf.  B. eximius), and the thoracic dorsum anteriorly black (cf.  B. festivus). Genitalia ( Fig. 182) with the gonostylus reduced to a transverse ‘S’-shaped band (cf. non-  rufipes-group), the proximal inner projection broadened distally and bifid (cf.  B. eximius); volsella projecting beyond gonostylus by ca 5 × its breadth at its midpoint (cf.  B. eximius), without an obvious inner distal process or hook (cf. non-  rufipes-group); penis valve head with the outer flange greatly expanded as a triangular projection (cf. non-  rufipes-group); eye unenlarged relative to female eye.    Material examined   Holotype   INDONESIA• ♀(worker), holotypeof  Bombus rufipes Lepeletier, 1835by monotypy; Java; OUMNH(examined PWfrom photographs). The original description refers to a single worker with no evidence that the author had additional specimens in front of him, so this worker can be recognised as the holotype by monotypy. Recognising this specimen as the holotype invalidates the neotype designation by Frison (1930: 11).  Material sequenced( 2 specimens)   INDONESIA• 2 ♂♂; East Java, Mt Argopuro; 7.9756° S, 113.5193° E;  5 Mar. 2010; BOLD seq: 1555E01, 1555E02; SK: ML249, ML250.    Global distribution (Southeast Asian mountain species primarily in the southern islands) Southeast Asia: MALAYSIA: Peninsular Malaysia, INDONESIA: Sumatra, Java,? Kalimantan,? Sulawesi. (NHMUK, NHMW, PW, OUMNH, SK.)  It was reported previously (by Williams in Starr 1989) that there is a male labelled “Bandjar/ Borneo” (= Banjar?) in the NHMUK collection (examined PW). The only other bumblebee species recorded apparently from Borneo is the holotypeof  B. folsomi( Frison, 1923), which is labelled “Kina Bala/ N. Borneo” (= Gunung Kinabalu, Sabah). However, the latter specimen appears to be a mislabelled queen of  B.( Pyrobombus) ephippiatusSay, 1837, probably originating from Costa Ricaor Panama( Starr 1989; Williams 1998, examined PW). The NHMUK “Bandjar/ Borneo” specimen of  B. rufipesreached this collection from B. Pittioni’s collection that was assembled in Austria(the specimen carries a det. label from F. Maidl, who was at Vienna’s NHMW museum, and a label “Pittioni Coll./Turner Bequest/ B.M. 1954–79”). The NHMW collection has a further three workers and two males with the same locality data in the same handwriting and with the same identification labels. Of nine localities named ‘Bandjar’ identified in Malaysiaand Indonesiafrom USmilitary gazetteers ( USBGN 1968, 1970), one is in Borneo and six are in Java. Frison (1930)interpreted ‘Bandjar’ as referring to Java for a queen, two workers, and the seven males of  B. rufipesloaned to him from the ‘Zoological Museum, Buitenzorg, Java’ (= Bogor) collection. Two further workers of  B. rufipeswithout locality labels are listed by Frison that were loaned to him by Maidl from the NHMW. Handlirsch (1891)wrote that all but one of the  B. rufipesthen in the NHMW collection were from eastern Java (for the other, see below). The occurrence of  B. rufipeson Borneo, perhaps on the higher Maratus mountains of south-eastern Kalimantan near Banjar (across the Java Sea from Java and Sumatra), remains possible, but would need to be confirmed. Two other records appear to extend the known range. One is a queen in the NHMUK from “ Selangor/ Bukit Kutu” (Peninsular Malaysia, examined). There is no obvious reason to doubt this record. Another is a queen in the NHMW collection labelled “N. Celebes/ 1908” (= northern Sulawesi, examined PW). The occurrence of  B. rufipesin northern Sulawesiwould need to be confirmed because it is ca 2000 kmfrom the nearest sites with confirmed records for  B. rufipeson Java(see the Discussion). The species is widely distributed but not common in collections.   Behaviour Some aspects of the behaviour of this species have been described ( Michener & Amir 1977; Kato et al.1992). 4989414301 OUMNH, PW Indonesia 48 49 Java 47 48 1 1 Jawa Timur holotype 2867606390 2010-03-05 Indonesia -7.9756 Mt Argopuro 7 1.1555 48 49 2 East Java