Bombus home | Introduction | BumblebeeID | Find species by Group, Region, Colour, A-Z | References


Find world species by region
click on a biogeographic region:


biogeographic regionsOrientalOrientalE NeotropicalE NeotropicalPalaearcticArcticArcticPalaearcticArcticPalaearcticE NeotropicalW NearcticW NearcticW NearcticW NearcticJapanJapanArcticArcticSumatraE NearcticN NeotropicalS NearcticS NearcticPalaearcticW NeotropicalW NeotropicalW NeotropicalS Neotropical
Biogeographic regions of the world (numbered) for bumblebees
(updated from Williams, 1996 [pdf] ).


  1 Sumatra 7 E Nearctic
  2 Oriental 8 S Nearctic
  3 Japan 9 N Neotropical
  4 Palaearctic 10 W Neotropical
  5 Arctic 11 E Neotropical
  6 W Nearctic 12 S Neotropical




Britain | Europe | USA+Canada | Introductions of non-native species


12 biogeographic regions

Biogeographic regions
Equal-area grid map
Bumble bee distribution data



Biogeographic regions

Rather than using the familiar traditional biogeographic regions that are based largely on birds (Sclater, 1858; Wallace, 1876, map below), the biogeographic regions used here are based directly on their bumblebee faunas.


Wallace map

Biogeographic regions of the world from Wallace (1876)


Biogeographic regions for bumblebees were derived using a TWINSPAN classification. This divides the equal-area (611,000 km) grid-cells (see below and the map of species richness) into 12 groups by differences in the species composition of their bumblebee faunas (updated from Williams, 1996 [pdf] ).


The resulting faunal regions for bumblebees are broadly similar to the traditional biogeographic regions based largely on birds. Where these regions correspond, the traditional names are used here. More regions are recognised in the New World because bumblebee species are fewer there and the species' distribution patterns are more concordant.


In contrast, the island bumblebee faunas of Japan and Sumatra do not agree with the traditional biogeographic regions. The Japanese and Sumatran Regions are distinguished from the neighbouring Oriental and Palaearctic Regions by having a subset of their species, with either a low proportion of endemics (Japan), or very few species in total (Sumatra). Similarly, most species recorded in the Arctic Region are shared with the more southerly boreal fauna, although species of the subgenus Alpinobombus are strongly associated with this region.



Equal-area grid map

The maps used here are based on a cylindrical, equal-area projection of the world (excluding Antarctica). However, this does not ensure equal areas of land among grid cells, or equal areas of habitat suitable for bumblebees.


The map shown above uses a cylindrical equal-area projection that is orthomorphic (minimum shape distortion) at 46° North and South (where bumble bee records are particularly plentiful). Intervals of 10 longitude (the map is opened at 30 West) are used to calculate intervals of latitude (4°59', 9°59', 15°03', 20°14', 25°36', 31°13', 37°12', 43°41', 51°00', 59°41', 71°44' North and South of the equator) that provide equal-area grid cells of approximately 611,000 km². A grid map shows the grid coordinates superimposed with the known extent of the indigenous distribution of bumblebees.



Bumblebee distribution data

This checklist was compiled in conjunction with distribution data for use in biogeographic studies. The maps of world-wide distribution were designed specifically for the coarse-scale comparisons of regional bumblebee faunas described above (Williams, 1996 [pdf], 1998 [pdf] ).


Aside from any difficulties in identifying species or localities, comparisons among faunas are complicated by two principal factors: first, by differences in sampling effort (as illustrated by 'species-accumulation curves', e.g. Colwell & Coddington, 1994); and second, by differences in the extent of sampling areas ('species-area effects', e.g. MacArthur & Wilson, 1967). Fortunately for the first problem, the attractiveness of bumblebees to collectors has ensured that they have been relatively intensively sampled, so that most faunas are relatively well known. But in order to reduce this problem further, rather than extrapolate local richness and lose information on individual species, the expected distributions of some species are interpolated on the basis of knowledge of their habitat associations. To reduce the second problem of species-area effects, equal-area grid cells are used.


Because the intention is to study biogeographic patterns, maps are required to show all historical records, including data from areas where species may now be extinct. On the other hand, data exclude fossil taxa (reviewed by Zeuner & Manning, 1976) and documented introductions.


Maps for every species are not included with this checklist because many data are still being collected and analysed. The sources of the distribution data will be presented in a later atlas.


For each subgenus a preliminary map of species richness is included as a general guide. These blue-scale maps use equal-interval classes, which have the advantage that each blue-scale class remains consistent in its range of richness values within each map. Scale keys are not included with each map because the numbers of species are shown directly on each grid cell. The maps were made using WORLDMAP software.


research publications | books on British bumblebees | links and credits


top top of page