The name Bombus flavifrons has been applied to two different bumble bee species in North America, one described by Ezra Cresson in 1863, the other by Frederick Smith in 1866. Bombus flavifrons Cresson, 1863 is currently considered the valid name for a common and widespread species in western North America, while B. flavifrons Smith, 1866 is considered a junior homonym of B. flavifrons Cresson, and thus not available. Consequently, B. columbicus Dalla Torre, 1890 was offered as a replacement name for B. flavifrons Smith. Smith’s taxon was described from a now-lost type specimen taken on Vancouver Island, British Columbia, Canada but has been considered a junior synonym of B. vosnesenskii Radoszkowski, 1862 since 1888. Prior to 2012 no other records of B. vosnesenskii were known from Vancouver Island, and it was considered very rare in Canada up until the early 2000s, known only from a few specimens from the southcentral mainland of British Columbia adjacent to the United States border. Bombus caliginosus (Frison, 1927) was not recorded in the literature from Canada until 2014 and is known only from five specimens collected in the 1970s at Victoria, Vancouver Island. Bombus caliginosus is distinguishable from B. vosnesenskii with some difficulty.

Due to the morphological similarity of B. vosnesenskii and B. caliginosus, and their historical scarcity and disjunct distributions in Canada, we propose that B. flavifrons Smith is conspecific with B. caliginosus, a species that was not described until almost 40 years after the synonymy of B. flavifrons Smith under B. vosnesenskii. As the type specimen of Smith’s taxon is seemingly lost, we also designate a neotype of B. flavifrons Smith from Vancouver Island to support the proposed nomenclature. Though Bombus columbicus is here considered a senior synonym of B. caliginosus syn. nov., under Article 23.9.1 of the ICZN Code, we designate the name nomen oblitum to maintain the prevailing use of B. caliginosus for this taxon.

Bombus caliginosus, Bombus vosnesenskii, bumble bees, extirpated species, nomenclature, North America, taxonomy

Bombus flavifrons Cresson, 1863, also known as the yellow-fronted or yellow head bumble bee, is a common species of western North America, ranging well into the western Arctic, with records from as far east as the Hudson’s Bay region of Ontario (Laverty and Harder 1988; Williams et al. 2014; but see Sheffield and Williams 2025). Frederick Smith (in Lord 1866) also used the name Bombus flavifrons for another yellow-faced North American species, describing B. flavifrons Smith, 1866 as being similar to Bombus californicus Smith, 1854 but differing in having darker wings, the face and vertex being pale yellow, and having a narrower yellow band on the metasoma (i.e. tergum 4). This specimen(s) was collected sometime between 1858–1862, during which time J.K. Lord was stationed on Vancouver Island, and it was the only new bee species described in that work, though four other bumble bee species were mentioned¹.

Handlirsch (1888) synonymized B. vosnesenskii Radoszkowski, 1862 (as Bombus Wosnesenskyi) and B. flavifrons Smith under B. californicus. However, the type specimen of B. californicus is a female, and is now considered by some (i.e., Williams et al. 2014; Sheffield and Heron 2019) to be a junior synonym of B. fervidus (Fabricius, 1798), while the male allotype was not conspecific and is consistent with concepts of B. vosnesenskii (see Franklin 1913). Though the common name of B. vosnesenskii should be Vosnesensky bumble bee (Williams et al. 2014), yellow-faced bumble bee has also been used (e.g., Fraser et al. 2012); this is consistent with Smith’s use of the species epithet flavifrons for this taxon. However, it seems unlikely that B. flavifrons Smith is a junior synonym of B. vosnesenskii as no older publications mentioning B. vosnesenskii from British Columbia (i.e., Scullen 1927; Hicks 1929; Buckell 1951; Stephen 1957; Thorp 1970) recorded it from Vancouver Island; in fact, only Buckell (1951) provided specific locality information for this species in Canada, recording just three specimens from Osoyoos from 1925. Though all these works document a long history of this species in the province, Scudder (1992, 1994) indicated that B. vosnesenskii was rare in the province and possibly deserving of conservation status assessment as even at that time it was known nationally from only Osoyoos (i.e., the specimens reported by Buckell in 1951). Since then, B. vosnesenskii has undergone rapid range expansion, and is now known to occur in the province from the central mainland to south coastal regions, including Vancouver Island (Fraser et al. 2012; Williams et al. 2014); it is possible that this is due, in part, to the use of commercially available colonies of B. vosnesenskii in British Columbia, though more recent natural spread into the province from the adjacent USA may also be a contributing factor. Oddly, a single specimen now at the Royal Saskatchewan Museum, Regina, Saskatchewan, Canada was found in a head of lettuce in a grocery store in New Minas, Nova Scotia, Canada. Our main purpose here is to resolve the nomenclature of Smith’s taxon based on the historic distribution of these species in Canada.

A wide range of taxonomic, faunistic, and ecological literature was mined for use of the names B. flavifrons, B. caliginosus, and B. vosnesenskii, and other names associated with these valid species; these were also searched for in the Biodiversity Heritage Library (BHL, https://www.biodiversitylibrary.org/). Descriptions, distribution, and other details linked to species epithets were reviewed; past synonymies were reviewed, and type material, or photos of types, were examined when possible.

All Canadian records of B. caliginosus (GBIF.org 2024a) and B. vosnesenskii (GBIF.org 2024b) were mined from the Global Biodiversity Information Facility (https://www.gbif.org/) to determine capture times and distribution history in Canada. In addition, over 400 specimens of B. vosnesenskii collected in Canada and held at the Royal Saskatchewan Museum and Royal British Columbia Museum, Victoria, British Columbia were examined and their identifications verified (using Williams et al. 2014, and characters discussed below). In addition, the specimen database and collections at The Natural History Museum, U.K. were searched for potential type material corresponding to Smith’s species. The known specimens of B. caliginosus from Canada were borrowed from the American Museum of Natural History, New York, to verify the identity of those occurrences. Lastly, the AMNH-Bee – Collaborative databasing of North American bee collections within a global informatics network project (Ascher 2025) and the Bumble bees of North America occurrence records database (Richardson 2025) were also utilized.

There is no evidence that B. vosnesenskii was found on Vancouver Island until the early 2000s (Fraser et al. 2012; Williams et al. 2014; GBIF.org 2024b). By contrast, in Canada B. caliginosus is only known from southernmost Vancouver Island (Williams et al. 2014; GBIF.org 2024a; Richardson 2025), these occurrences are based on five female specimens in the American Museum of Natural History. Bombus caliginosus is also known from adjacent Stuart Island in the San Juan Islands, and the Olympic Peninsula, Washington, USA.

Though there is currently no type specimen of B. flavifrons Smith to examine, we believe that there is a high probability that B. caliginosus is conspecific with B. columbicus so we designate a neotype specimen from Vancouver Island to fix the name to a specimen. Bombus columbicus is thus considered a senior synonym of B. caliginosus but based on Article 23.9.1 of the Code (ICZN 1999), the prevailing usage must be maintained if the senior synonym (i.e., B. columbicus) has not been used as a valid name after 1899, which is the case as B. flavifrons Smith was already, albeit seemingly incorrectly, considered a synonym of B. vosnesenskii at the time Dalla Torre (1890) proposed B. columbicus as a replacement name, so the conditions of Article 23.9.1.1 are met. The name B. caliginosus has subsequently appeared in numerous taxonomic and faunistic papers (e.g., Scullen 1927; Frison 1929; Burks 1951; Stephen 1957; Krombein 1967; Thorp et al. 1983; Dobson 1993; Williams 1998; Cameron et al. 2007; Fraser et al. 2012; Koch et al. 2012, 2015; Williams et al. 2014; Sheffield and Heron 2019; Fisher et al. 2022; Bartholomew et al. 2024), in addition to many ecological studies (e.g., Pitelka 1954; Hicks 1959; Moldenke 1976; Ackerman 1981; Patten et al. 1993; Knops and Barthell 1996; McFrederick and LeBuhn 2006; Rao and Stephen 2010; Kissinger et al. 2011; Thomson 2016; Strange and Tripodi 2019; MacPhail et al. 2020; Spiesman et al. 2021), so the conditions of Article 23.9.1.2 are also met.

 Bombus caliginosus (Frison, 1927), nomen protectum

Bombus flavifrons Smith, 1866, in Lord, 1866: 343 (appendix) [♀]. Preoccupied, not Bombus flavifrons Cresson, 1863. Neotype: ♀ (designated here; Fig. 1). Canada, British Columbia, Vancouver Island, Victoria [Island], Butchart Gardens, 30 May 1973, by S. Gittelman [American Museum of Natural History, no. AMNH_BEE 00120464].

Bombus columbicus Dalla Torre, 1890: 139. Replacement name for Bombus flavifrons Smith, 1866. Nomen oblitum.

Bremus caliginosus Frison, 1927: 376 [♂]. Syn. nov. Holotype: ♂. USA, California, Arcata, 16 September 1920, by C.D. Duncan [Illinois Natural History Survey, no. 179514].

Bremus caliginosus var. tardus Frison, 1927: 380 [♂]. Synonymy of Bremus caliginosus Frison, 1927 by Stephen (1957: 100). Holotype: ♂. USA, California, Carmel, 20 July 1921, by L.S. Slevin [California Academy of Sciences no. 2438].

Remark.

The original description of Bombus flavifrons Smith was of a female with a body length of 8½ lines (ca 18 mm), consistent with the neotype selected here (Fig. 1).

Figure 1. 

Neotype of Bombus flavifrons Smith, 1866 [= Bombus caliginosus (Frison, 1927)]; female, worker [American Museum of Natural History]. A. Lateral (with associated specimen labels) and B. Dorsal habitus; C. Face, and D. Metasoma in lateral view showing the posterior pale hairs of sternum 4 (red arrow).

Diagnosis.

Females of B. caliginosus have a longer malar space (length:width = 0.90, Fig. 2a) than those of B. vosnesenskii (length:width = 0.71, Fig. 2b) and in many specimens sternum 4 of the former has a weak apical (posterior) band of pale hairs, at least laterally (Thorp et al. 1983; Williams et al. 2014) (Fig. 1d), though as cautioned by Williams et al. (2014) this is also true for some specimens of B. vosnesenskii. Williams et al. (2014) also noted that tergum 4 of B. caliginosus often has some dark hairs basally, especially medially, while in B. vosnesenskii the black hairs are lacking, or few. Stephen (1957) noted that the relative lengths of flagellomeres 1 and 3 also differed, flagellomere 1 longer than 3 (3:2) in B. caliginosus, but the lengths are subequal in B. vosnesenskii. Williams et al. (2014) made note of the size differences of punctures on the apical margin of the clypeus, being small in B. caliginosus, but large in B. vosnesenskii.

Figure 2. 

Malar space of females of A. Bombus caliginosus (Frison, 1927) and B. B. vosnesenskii Radoszkowski, 1862. White horizontal lines show the width of the malar space or distance between mandibular hinges; yellow vertical lines perpendicular to white horizonal lines show the height of the malar space from the lowest margin of the compound eye to the lower edge of malar space between the mandibular hinges); the parallel vertical yellow and white lines to the left of each figure are the same lengths as those measuring the malar space, and are added to facilitate visual comparison.

Males of the two species can be also distinguished by the relative lengths of flagellomeres 1 and 3, the lengths being subequal in B. caliginosus, but flagellomere 1 shorter than 3 in B. vosnesenskii (Stephen 1957; Thorp et al. 1983). The males also differ in the shape of the apical recurved portion of the penis-valve head, narrowing apically in B. vosnesenskii, but remaining subparallel in B. caliginosus (Stephen 1957; Williams et al. 2014). Thorp et al. (1983) also provided illustrations noting differences in the shapes of sterna 7 and 8, the latter gradually narrowing apically (i.e., triangular) in B. caliginosus (their fig. 63), while widely truncate apically in B. vosnesenskii, with the lateral margins subparallel to base of process (their fig. 72).

Stephen (1957), Thorp et al. (1983), and Williams et al. (2014) have all stressed the difficulty of distinguishing B. vosnesenskii from B. caliginosus, especially for the females. Thorp (1969) even indicated that some of the specimens of B. caliginosus identified in the revision by Stephen (1957) were actually B. vandykei. Despite the morphological similarity, Thorp (1969) included B. caliginosus and B. vandykei, but not B. vosnesenskii, in his Flavifrons species group of the subgenus Pyrobombus. The distinctiveness of this group was supported in the molecular phylogenies of Hines et al. (2006), Cameron et al. (2007), and Hines (2008).

Bombus caliginosus is very rare in Canada, known only from the five confirmed records from a single collecting event at Victoria, British Columbia (and see Williams et al. 2014). By contrast, B. vosnesenskii, though also once rare in Canada (Scudder 1992, 1994) and restricted in distribution to the Osoyoos area of southcentral British Columbia (Buckell 1951), with three additional specimens collected in 1909 now known from Revelstoke (Richardson 2025), has become one of the most common bumble bees in southern parts of the province, extending its range onto Vancouver Island (Fraser et al. 2012). Although online sites such as Bumble Bee Watch (www.bumblebeewatch.org) and iNaturalist (https://www.inaturalist.org/) typically offer excellent occurrence documentation for bumble bees, these two species are morphologically similar enough on a macro scale that identification from the colour patterns shown on photographs are likely not reliable as there are no consistent differences (Williams et al. 2014). Identification may only be possible when the images show malar space length and/or in when the ventral surface of the metasoma is visible; males of these two species are even harder to reliably identify from photos. Though all 400+ specimens examined from British Columbia for this study have been identified as B. vosnesenskii, continued surveys and collection of specimens will be required to confirm the presence of B. caliginosus in, or its extirpation from, Canada.

Thanks to Dr. James Carpenter, American Museum of Natural History, New York, NY, for the loan of specimens of Bombus caliginosus from Vancouver Island, and Claudia Copely, Royal British Columbia Museum, Victoria, BC, for several loans of bumble bees from British Columbia. Thanks also to Doug Yanega, University of California, Riverside, CA, for past discussions on type materials and the ICZN Code. Thanks also to Leif Richardson (Xerces Society, Portland, OR) and John S. Ascher (National University of Singapore, Singapore) for helpful discussion and information on the historic distribution of these species in Canada, the identifications confirmed by John S. Ascher. We also thank the reviewers and the editor for helpful comments on the manuscript.