PART II - The Evolutionary Tree of Bees

5. The Seven Families: Mapping the Bee Lineage

Figure 2. Phylogeny of bees based on molecular and morphological data Phylogenetic relationships among major bee lineages based on multi-gene analyses, illustrating early divergence within Anthophila and the distinction between short-tongued and long-tongued bee clades. In this framework, groups traditionally treated as subfamilies (e.g., Dasypodainae, Meganomiinae, Melittinae) are considered at family level, reflecting their deep evolutionary separation. Source: Adapted from Danforth et al. (2006) via ResearchGate figure pageOriginal caption (adapted): “Phylogeny of bees according to Danforth et al. (2006). Subfamilies of Melittidae treated as family-level groups. LT = long-tongued bees.”

If the word bee collapses a vast diversity into a single familiar image, then the concept of families restores that lost complexity. The seven recognised families of bees are not merely taxonomic conveniences, but deep evolutionary divisions - branches that diverged early in the history of the Anthophila and have since followed independent trajectories for tens of millions of years.

To understand bees properly, one must begin to think in terms of these lineages, because each family represents not simply a group of related species, but a distinct evolutionary pathway - a different solution to the shared challenge of life as a pollen-dependent insect.

Modern phylogenetic work, drawing on both morphological datasets and increasingly robust molecular analyses, consistently resolves bees into seven principal families: Andrenidae, Halictidae, Colletidae, Melittidae, Megachilidae, Apidae, and the geographically restricted Stenotritidae (Danforth et al., 2006; Hedtke et al., 2013; Almeida et al., 2023). These divisions reflect ancient divergences that likely occurred relatively early in bee evolution, shortly after the origin of the clade itself (Cardinal & Danforth, 2013).

Crucially, these families are not variations on a single theme. They represent independent evolutionary experiments, each exploring different ecological and behavioural strategies. Some lineages diversified extensively, radiating into a wide range of niches, while others remained more constrained, retaining specialised or geographically restricted distributions. What unites them is not similarity, but shared ancestry; what distinguishes them is the direction in which that ancestry has been taken.

There is also an asymmetry within this structure that is worth emphasising. Families such as Apidae and Halictidae are exceptionally diverse, encompassing thousands of species and a wide range of behavioural strategies - from solitary nesting to complex eusocial systems with communication and division of labour. In contrast, families such as Melittidae or Stenotritidae are comparatively small and often highly specialised, with restricted geographic distributions.

Yet their importance is not diminished by their size.

On the contrary, these smaller families often preserve traits that are evolutionarily informative, offering insight into earlier stages of bee diversification and helping to resolve phylogenetic relationships across the clade (Danforth et al., 2006). In this sense, the bee phylogeny is not simply a branching diagram, but a historical record, in which different lineages retain different elements of ancestral biology.

To think in terms of families, then, is to shift perspective.

The honeybee - so dominant in cultural and scientific narratives - ceases to be the centre of the story and becomes instead a member of one branch. The focus moves outward, from a single species to the structure of the lineage itself.

And with that shift, the central question changes.

It is no longer what is a bee?

but:

How many fundamentally different ways are there to be a bee?

6. Deep Time: Divergence and Early Evolutionary Splits

The origin of bees lies deep in geological time, within ecosystems that would be almost unrecognisable today. During the mid-Cretaceous period, approximately 100–120 million years ago, flowering plants (angiosperms) underwent a rapid and transformative radiation. Landscapes previously dominated by gymnosperms became increasingly structured by flowering plants, whose reproductive systems introduced new ecological opportunities in the form of nectar and pollen (Grimaldi, 1999; Danforth et al., 2006).

It was within this context that certain lineages of predatory apoid wasps began to shift their ecological strategy.

Rather than provisioning their larvae exclusively with paralysed arthropod prey, these ancestral insects began to incorporate pollen into their reproductive cycle. Over time, this transition became more consistent and ultimately obligatory, marking the emergence of the earliest bees (Cardinal & Danforth, 2013).

This shift was neither instantaneous nor inevitable. It required a coordinated suite of changes - behavioural, morphological, and physiological - that collectively redefined the organism. Early bees were not simply wasps that visited flowers more frequently; they were organisms undergoing a fundamental ecological transformation.

Natural selection favoured individuals that could more efficiently locate flowers, extract nectar, collect pollen, and transport it effectively. Over evolutionary time: body hairs became increasingly branched and specialised for pollen retention, mouthparts adapted to exploit diverse floral morphologies, sensory systems tuned to floral colour, scent, and reward gradients, life cycles synchronised with plant phenology.

Once bees became fully dependent on pollen and nectar for reproduction, the conditions for diversification were established.

Molecular phylogenetic analyses indicate that major bee lineages diverged relatively early following this transition, with family-level splits occurring soon after the origin of the clade (Hedtke et al., 2013). What followed was not a linear progression, but a branching radiation, in which different lineages explored distinct ecological strategies.

It is tempting to describe some of these lineages as “primitive” and others as “advanced,” but such terminology is misleading. All extant bee families have undergone continuous evolution over immense spans of time. What differs is not their level of development, but the direction of their specialisation.

Some lineages retained solitary, ground-nesting lifestyles; others evolved complex social systems. Some became highly specialised in their floral associations; others remained generalists. These are not stages along a ladder, but trajectories across an evolving ecological landscape.

In this sense, early bee diversification can be understood as a process of ecological partitioning across deep time. As flowering plants diversified, so too did the strategies for exploiting them. Each lineage diverged by occupying a slightly different niche - defined by morphology, behaviour, timing, or environment.

Over millions of years, these differences accumulated, producing the structured diversity observed today.

7. The Structure of Diversity: Why So Many Lineages Exist

The sheer number of bee species invites an obvious question: why are there so many?

The answer lies not in a single cause, but in the interaction of several reinforcing evolutionary processes, each contributing to the generation and persistence of diversity.

One of the most important of these is ecological partitioning. In any given environment, floral resources are not uniform. Different plant species bloom at different times, offer different rewards, and present different morphological challenges. Bees have diversified to exploit these variations.

Some species are active early in the season, others later. Some forage in cooler conditions, while others require higher temperatures. Some specialise on open, accessible flowers, while others are adapted to deep or structurally complex floral forms. Even within a single habitat, numerous bee species may coexist by occupying subtly different niches, reducing direct competition and allowing multiple lineages to persist simultaneously (Ollerton et al., 2011).

Closely linked to this is morphological diversification. The physical structure of a bee - its size, tongue length, hair distribution, and pollen-collecting apparatus - directly influences the range of floral resources it can exploit. These traits evolve in response to ecological pressures.

Long-tongued bees, for example, are capable of accessing nectar in flowers that are inaccessible to short-tongued species, effectively partitioning the floral resource space. Similarly, differences in pollen transport structures - such as scopal hairs or corbiculae - reflect alternative strategies for gathering and handling pollen. These variations contribute to functional diversity within pollination systems and enhance ecological resilience (Klein et al., 2007).

Behavioural flexibility introduces an additional layer of complexity. In certain lineages, particularly within the Halictidae, social organisation is not fixed but variable. Species may exhibit solitary, communal, or primitively eusocial behaviour depending on environmental conditions (Gibbs et al., 2012). This plasticity allows populations to respond dynamically to ecological pressures and may facilitate both short-term persistence and long-term diversification.

Perhaps the most significant driver of bee diversity, however, is their relationship with flowering plants.

Bees and angiosperms are engaged in an ongoing coevolutionary interaction, in which changes in one group influence the evolutionary trajectory of the other. Flowers evolve traits that favour specific pollinators, while bees evolve traits that allow them to exploit particular floral resources more efficiently. In some cases, this leads to extreme specialisation, with bee species becoming tightly associated with specific plant taxa.

These relationships can be so precise that the loss of one partner directly impacts the survival of the other (Ollerton et al., 2011).

Taken together, these processes - ecological partitioning, morphological innovation, behavioural flexibility, and coevolution - generate a system in which diversity is not only produced, but stabilised over time.

Bee diversity is therefore not a random accumulation of species, but a structured outcome of long-term evolutionary dynamics.

8. Families as Evolutionary Worlds

By this point, the concept of bee families begins to shift in meaning. They are no longer abstract taxonomic categories, but distinct evolutionary worlds, each defined by its own constraints, adaptations, and ecological strategies.

To move from one family to another is not merely to encounter different species, but to enter a different mode of existence.

Within the Andrenidae, one finds predominantly ground-nesting species, often emerging in close synchrony with seasonal flowering events, their life cycles tightly linked to soil conditions and climatic timing. In the Halictidae, diversity becomes more fluid, with metallic forms and a remarkable range of social behaviours, from solitary nesting to eusocial organisation.

The Megachilidae introduce a different suite of innovations, centred on the use of external materials - leaves, mud, and resin - in nest construction, along with the distinctive use of abdominal scopae for pollen transport. The Apidae, perhaps the most familiar family, encompass both the highly organised societies of honeybees and bumblebees and the more solitary or specialised behaviours of carpenter bees and orchid bees.

What becomes clear, when these families are considered together, is that there is no single trajectory of bee evolution.

Instead, there are multiple trajectories, each shaped by different combinations of ecological opportunity and constraint. Some lineages have evolved increasing social complexity; others have remained solitary. Some have become generalists; others have specialised intensely. Some exhibit elaborate communication systems; others rely on simpler, yet equally effective, strategies.

This is the point at which the concept of bee diversity becomes fully realised.

It is not simply that there are many bees.

It is that there are many fundamentally different ways of being a bee.

And it is from this branching structure - from these seven evolutionary worlds - that the rest of the story unfolds.

Next in the series - Part III: The First Lineages

If Part II mapped the branches of the bee tree, the next step is to walk along them. In Part III, we move from structure to substance - entering the earliest bee families in detail, examining how each lineage lives, nests, feeds, and evolves. Here, the abstraction of “families” gives way to real organisms, real behaviours, and real ecological roles - revealing just how different one bee can be from another.

References

Almeida, E. A. B., et al. (2023). Advances in bee phylogeny and classification. Annual Review of Entomology.

Cardinal, S., & Danforth, B. N. (2013). Bees diversified in the age of eudicots. Proceedings of the Royal Society B, 280, 20122686.

Danforth, B. N., Sipes, S., Fang, J., & Brady, S. G. (2006). The history of early bee diversification. PNAS, 103(41), 15118–15123.

Gibbs, J., et al. (2012). Facultative eusociality in bees. Biological Reviews.

Grimaldi, D. (1999). Co-radiation of pollinating insects and angiosperms. Annals of the Missouri Botanical Garden, 86(2), 373–406.

Hedtke, S. M., Patiny, S., & Danforth, B. N. (2013). The bee tree of life. Molecular Phylogenetics and Evolution, 69, 348–358.

Klein, A. M., et al. (2007). Pollinators in global agriculture. Proceedings of the Royal Society B, 274, 303–313.

Ollerton, J., Winfree, R., & Tarrant, S. (2011). How many flowering plants are pollinated by animals? Oikos, 120, 321–326.