Bombus (Alpigenobombus) grahami ( Frison, 1933 )

Williams, Paul H., An, Jiandong, Dorji, Phurpa, Huang, Jiaxing, Jaffar, Saleem, Japoshvili, George, Narah, Jaya, Ren, Zongxin, Streinzer, Martin, Thanoosing, Chawatat, Tian, Li & Orr, Michael C., 2023, Bumblebees with big teeth: revising the subgenus Alpigenobombus with the good, the bad and the ugly of numts (Hymenoptera: Apidae), European Journal of Taxonomy 892 (1), pp. 1-65 : 45-47

10. Bombus (Alpigenobombus) grahami ( Frison, 1933)

Figs 9 , 94‒103 , 115

Bremus (Alpigenobombus) grahami Frison, 1933 [[30] September]: 334. Note 1.

Alpigenobombus beresovskii Skorikov, 1933a [30 September: Pesenko & Astafurova 2003]: 248. Syn. nov.

Bombus Alpigenobombus grahami [subsp.] melani Wang & Yao, 1993: 422 . Syn. nov.

Alpigenobombus (Alpigenobombus) beresovskii ‒ Skorikov 1923: 156, published without description or indication, unavailable name.

Alpigenobombus berezovskii ‒ Skorikov 1931: 204, published without description or indication, unavailable name.

Bremus (Alpigenobombus) grahami var. [not subsp.] ordinatus ‒ Frison 1935: 353, infrasubspecific.

NOTE 1. Acceptance of B. grahami as the valid name in preference to B. beresovskii follows Williams (2022a) as the First Reviser ( ICZN 1999: Article 24). Bombus grahami is the name in most widespread use for this species in the literature.

Species-taxon concept and variation

The taxon concept of the species B. grahami here is modified from the long-standing interpretation (Williams 2022a), in that it includes the taxon beresovskii , based on: (1) our PTP analysis supports independent species-level coalescents in the COI gene ( Fig. 12 ); corroborated by (2) diagnostic morphological character states (see the keys).

The PTP and morphological results ( Fig. 12 , keys) support the interpretation that the divergent colour patterns of the taxa grahami s. str. and beresovskii are conspecific (Williams 2022a). The available COIbarcode-like sequences may all be low-divergence neonumts ( Fig. 11 ). From morphology, the taxon melani (holotype queen Fig. 99) is closely similar to the taxon beresovskii (syntype queen Fig. 100).

For the typically coloured individuals of B. grahami , the wings of the males are (unusually among bumblebees) much paler than the wings of the females. The wings are also slightly paler for females of the taxon beresovskii (males not seen).

Some females with the beresovskii colour pattern (from northern Sichuan: Figs 99–100) have a dark colour pattern that resembles B. angustus , but others (from Shaanxi: Fig. 98) have a light colour pattern that resembles B. validus . One worker (Fig. 98) has a distinct band of grey hairs fringing the thoracic dorsum anteriorly and posteriorly with some grey hairs along the midline. Males of the taxa beresovskii and melani are unknown to us.

Bombus grahami females, with the grey-thorax colour pattern in the west (Himalaya) and centre (Hengduan) of its range (Figs 94‒97), appear to mimic the commoner B. (Orientalibombus) funerarius Smith, 1852 ( Williams 2007: fig. 5g). In contrast, the more extensively black-thorax colour pattern in the north-east (in the hills to the north of the Sichuan Basin: Figs 98‒100) appears to mimic the abundant B. (Melanobombus) pyrosoma ( Williams 2007: close to fig. 5m). Some of these females (Fig. 98) resemble closely B. validus (Figs 70‒71) in colour pattern, but can be distinguished easily by the shorter oculo-malar distance of B. grahami .

Type material

Bremus (Alpigenobombus) grahami Frison, 1933 [[30] September]: 334. Holotype by original designation: ♀ (worker) Sichuan, China ( USNM). Examined .

Alpigenobombus beresovskii Skorikov, 1933a [30 September: Pesenko & Astafurova 2003]: 248.

Syntype (unpublished Podbolotskaya lectotype): ♀ (queen) Sichuan, China ( ZIN). Examined .

Bombus (Alpigenobombus) grahami [subsp.] melani Wang & Yao, 1993: 422 .

Holotype by original designation: ♀ (queen) Henan, China ( IOZ). Examined.

Morphological diagnosis

Female

Wings very darkly clouded with brown with the veins dark brown, hair short, oculo-malar area shorter than broad, clypeus in its central area with only a few large and small widely-spaced punctures (cf. B. breviceps ); hair of the thoracic dorsum black, either usually with many grey hairs intermixed so as to appear silvery olive-grey, with the majority of the hair along the longitudinal midline white, or if entirely black, then T3 predominantly orange-red.

Male

Wings nearly clear with the veins dark brown, hair short; genitalia ( Fig. 115 ) with the gonostylus half as long on its outer side than on its inner side with the distal lobe slightly reduced and broadly flattened, appearing as a triangular pointed inwardly-projecting process only from the inner aspect (cf. B. genalis , B. breviceps ); hair of the thoracic dorsum yellow, often between the wing bases with many black hairs intermixed.

Material sequenced in Fig. 12

CHINA • 1 ♀ (queen); Sichuan, Taoyuan ; 32.6924° N, 106.8336° E; 21 Sep. 2011; PW seq: PWB01; IAR: AG#001 • 1 ♀ (worker); Yunnan, Lijiang ; 27.0156° N, 100.1714° E; 12 Jul. 2017; H. Liang leg.; KIB seq: KIB030 ; KIB: AG#076 .

Additional sequences in Fig. 10 and haplotype duplicates

BHUTAN • 1 ♀ (worker); Thimpu, Dochu-La; [27.4898° N, 89.7504° E]; 7 Jul. 1996; L. Blommers leg.; BOLD seq: 6880H10; RMNH: AG#094 .

CHINA • 2 ♀♀ (workers); Yunnan, Lijiang ; 27.0156° N, 100.1714° E; 2 Aug. 2017; H. Liang leg.; KIB seq: KIB092 , KIB097 ; KIB: AG#075 , AG#077 • 2 ♀♀ (workers); same collection data as for preceding but 12 Aug. 2017; KIB seq: KIB116 , KIB032 ; KIB: AG#078 , AG#079 • 1 ♀ (worker); Sichuan, Meigu ; 28.329° N, 103.129° E; GenBank seq: FJ175354; SC: AG#096 • Yunnan, Yulong Snow Mountain ; [27.047° N, 100.261° E]; H. Liang leg.; GenBank seq: MT906009; KIB: AG#150 .

INDIA • 1 ♀ (worker); Arunachal Pradesh, 1300 m E of Lama Tukkut ; 28.9949° N, 95.2926° E; 30 Aug. 2017; NCBS seq: BE782; NCBS: AG#203 .

NEPAL • 1 ♁; Janakpur, Dongo Kharka ; 27.9022° N, 86.2865° E; 23 Aug. 1983; I. Kudo leg.; BOLD seq: 1551H05; SEHU: AG#119 .

Global distribution

Himalaya, Hengduan, and into Central China: Nepal, India (Sikkim, Arunachal Pradesh), China (Xizang, Yunnan, Sichuan, Gansu, Shaanxi, Chongqing, Hubei, Henan, Guizhou): IAR, IOZ, KIB, NCBS, NHMUK, PW, RMNH, SEHU, USNM, ZIN.

This species is recorded at elevations of 2650‒2800 m in the Himalaya ( Williams et al. 2010; Streinzer et al. 2019), and 858‒3040 m in the Hengduan and in Shaanxi ( Williams et al. 2009; An et al. 2014). Bombus grahami is active in open areas in the middle and upper forest zones, but also inside the forest. Individuals with a black-thorax colour pattern (taxon beresovskii ) occur in the hills to the north of the Sichuan basin, where they appear to be extremely rare (only three queens and three workers have been examined: IAR, IOZ, PW) and are not known to co-occur with individuals with the olive-thorax colour pattern (although many more data are needed to clarify the situation).

Behaviour

Male eye slightly enlarged relative to female eye, male mate-searching behaviour unknown.

Figures

Figs 1‒9. Individuals of the subgenus Alpigenobombus Skorikov, 1914 (with photo credits). 1. Bombus mastrucatus Gerstaecker, 1869, worker Norway robbing (P. Haringsma). 2. B. kashmirensis Friese, 1909 s. str., worker China-Sichuan robbing (PW). 3. B. kashmirensis (taxon meinertzhageni Richards, 1928) worker India-Kashmir-Zanskar (PW).4. B. sikkimiFriese, 1918, worker India-Arunachal (MS). 5. B. nobilis Friese, 1905 s. str., worker China-Yunnan (ZR). 6. B. genalis Friese, 1918, worker India-Arunachal (MS). 7. B. breviceps Smith, 1852 (taxon dentatus Handlirsch, 1888) worker China- Yunnan (PW). 8. B. breviceps (taxon channicus Gribodo, 1892) worker Thailand (CT). 9. B. grahami (Frison, 1933) worker India-Arunachal (MS). Some images reversed.

Figs 19–105 (see pages 28–29). Simplified diagrams for the colour patterns of the hair on the dorsum for particular female (f) and male (m) specimens of the species from Fig. 12. The dorsum is divided into regions, each of which shows only the predominant or most apparent colour for that region, using a simplified colour palette, with olive indicating a mixture of black and yellow hair, and grey indicating a mixture of black and white hair.

Figs 106–115. Morphology of the male genitalia for species of the subgenus Alpigenobombus Skorikov, 1914, from the dorsal aspect, anterior at the bottom of the image, posterior at the top. 106. Bombus wurflenii Radoszkowski, 1860, Turkey. 107. B mastrucatus Gerstaecker, 1869, Austria. 108. B. kashmirensis Friese, 1909, India-Kashmir. 109. B. rainai Williams, 2022, India-Kashmir. 110. B. sikkimi Friese, 1918, Nepal. 111. B. nobilis Friese, 1905, China-Sichuan. 112. B. validus Friese, 1905, China-Gansu. 113. B. genalis Friese, 1918, China-Yunnan. 114. B. breviceps Smith, 1852, ChinaSichuan. 115. B. grahami (Frison, 1933) China-Sichuan (left penis-valve recurved hook missing). Scale bars = 1 mm.

Fig. 12. Interpretation of filtered MrBayes estimate of phylogeny as a metric tree (outgroup B. festivus Smith, 1861, not shown) for the subgenus Alpigenobombus Skorikov, 1914, from COI-like sequences from the sequences identified in black and orange in Fig. 11 together with Bayesian Poisson-tree-process (PTP) models re-applied for assessing support for species’ gene coalescents by maximum likelihood (for the 11 most likely candidate species). The scale bar is calibrated in substitutions per nucleotide site. Sequence labels and branch colours as in Fig. 10. Numbers above nodes are the Bayesian support values that all daughter haplotypes are parts of a single species. Grey spots show likely mitochondrial-to-nuclear transfers of the low-divergence numts accepted for estimating this tree. Asterisks mark sequences used as informal proxies for the type specimens of each of the taxon names in Table 2. To the right in grey are shown the interpretations of the PTP results as candidate species using the oldest available names for the species.

Fig. 11. Re-interpretation of sequences from Fig. 10, re-drawn with the percentage of nucleotides A and T at codon position 3 (AT3%) for each sequence given after the sequence label. Sequences are classified: (a) in purple – earlier-diverging groups of sequences without unique morphology that often have lower AT3% and shorter sequences that appear to duplicate species as pseudospecies in Fig. 10 – these sequences are interpreted as older, more divergent numts; (b) red – remaining sequences with low AT3% (<95%) – these are interpreted as younger, less divergent numts; (c) orange – sequences from the red group but with the highest AT3% available for some taxa, selected to retain these taxa in the analysis; (d) grey and brown – sequences that are duplicate haplotypes for the black and orange sequences above and often shorter; (e) black – likely orthologous sequences with high AT3% (≥ 95%) and unique haplotypes.

Fig. 10. Default interpretation of MrBayes estimate of phylogeny as a metric tree (outgroup B. festivus Smith, 1861, not shown) for the subgenus Alpigenobombus Skorikov, 1914, from COI-like sequences from GenBank and BOLD databases with additions from the authors, analysed with Bayesian Poissontree-process (PTP) models for assessing support for species’ gene coalescents by maximum likelihood (PTP scores are shown above branches: scores approaching 1, and where branches change from blue to red, indicates where the most likely species’ gene coalescents are identified for 17 candidate species). The scale bar is calibrated in substitutions per nucleotide site. Each sequence is labelled with: sequence length; a morphological taxon name; a code consisting of a sequence identifier from the project database and a specimen identifier from the online database; its country and (for larger countries) state or province of origin.