“If Honeybees Descended from Wasps, Why do we Still Have so Many Wasps?”
Reader John asked this in response to bees eating their larvae. From a human perspective, bees are not “vegetarian”. Indeed, diseased and weak young larvae is eaten to recycle nutrients and prevent the infection of other brood cells. Before there were bees, there were wasps, and before there were wasps, there were predatory insects. Bees are one branch of the wasp family, for like them, we have solitary, social and parasitic bees.
What evidence shows that bees descended from wasps? This article examines the genetic, anatomical (“morphological”), and fossil data for the evolution of bees from wasps. I had already covered what bees could evolve into, so think of this as an introduction to “what bees evolved from”.
Genetic Evidence for Wasp Ancestry
Certain regions of the DNA sequences are highly conserved between species – think in sequences for embryo development (Hox genes) or cell division (Wnt & Notch signalling genes), or they do nothing at all (“non-coding regions”). Branstetter and colleagues used these ultra-conserved regions from fresh and museum-preserved specimens to determine the evolutionary relationships between the Hymenopteran families (Bransetter et al., 2017).
They showed that sawflies were the oldest lineage, appearing over 250 million years ago (“mya”) in the late Permian period. This was followed by the emergence of parasitic wasps in the early Jurassic, with both species making up the suborder “Apocrita”; the “narrow-waisted hymenopterans”. The first stinging wasps and perhaps first ants occurred later in the Jurassic within the suborder Aculeata, followed by stingless bees and crabronid wasps in the early Cretaceous. During the middle Cretaceous 125 mya, the bee families emerged- Perhaps owing to the spread and dominance of flowering plants (Almeida et al., 2023)?
Mitochondrial DNA Sequences Show Further Evidence of Relativity
Mitochondrial genomes are relatively small. They are inherited from the mother, rarely recombine (unlike nucellus DNA), and have a high evolutionary rate. The gene content of animal mitochondria is constant and is typically conserved within lineages; any change can be picked up and the respective species identified. If none of this made sense to you, it’s enough to know that mitochondrial DNA can be used to establish evolutionary relationships.
Zheng and colleagues used this mitochondrial genome to show relatedness between bees and wasps in the Crabronidae family (Zheng et al., 2018). Further relationships were determined within the families of long-tongued (e.g. honey bees, stingless bees) & short-tongued bees (e.g. the mining, sweat and plasterer bees).
Fossils Show Primitive Wasps with Pollen
Two fossils, Prosphex anthophilos and Melittosphex burmensis, are thought to be the first pollen-feeding wasps during the Cretaceous period, 100 mya.
Both fossils were extracted from Burmese amber, and micro-CT scans showed pollen grains in its mouth and inside its oral cavity of P. anthophilos (Grimaldi et al., 2019). Based on both direct evidence (an insect fossil with pollen) and indirect evidence (i.e. insect lineages that are pollinators today or insects that have floral-feeding features like a long proboscis), the authors suggested that P. anthophilos is an early bee.
M. burmensis showed hairs on its body, which may have trapped pollen grains. However, the authors take this with a grain of salt, as the “hairs are barely plumose, it [M. burmensis] lacks the pronotal lobes [“solid little pimple” on the thorax] typical of apoids, and it has a broad pronotum [a cover on the top part of the thorax], like chrysidoids [parasitoid wasps]” (Grimaldi et la., 2019). Indirectly, these wasps and their families could have been pollinators due to their strong flight, intelligence (relative to other insects) and stinging capabilities. The sting is a significant adaptation, as the insect can now defend itself while foraging on exposed flowers. So successful are there adaptations that a whole genus arose to mimic them: the hover flies.
Danforth & Poinar suggest these early wasp reveal a transitional morphology between lineages, bridging the gap between surviving wasps and modern bees (Danforth & Poinar, 2011). However, the sample collected was a male, making it difficult to confirm if Melittosphex was a true pollen collector (i.e. using it feed to her larvae). The authors state that even if a female fossil was found, the verdict remains inconclusive, as some modern, non-parasitic bee groups carry pollen internally (e.g. Euglossinae & Hylaeinae families). Nonetheless, the hairs, wing venation, and leg anatomy on Melittosphex differ from the ancient wasps, supporting the hypothesis that this species was a primitive, pollen-collecting, ground-nesting bee.
Morphology: “Wasp Waist”
How do you tell a bee apart from a fly? You could describe the black and yellow colourations, but that doesn’t include spectrum of solitary and non-honey bees, nor does it account for their hoverfly mimics. You could also look at their body and observe the “constriction” between the abdomen and thorax. That’s a good start and puts you ahead of 98% of the population.
The remaining 2% of identification is for the specialists: bees have branched hairs on their body, wasps are generally hairless. Some bees carry pollen in the corbiculae on their hind legs, others carry it on the underside of their abdomen or in their internal crop. That’s not to say pollen wasps don’t exist – there are “vegetarian” wasp species, who are solitary and live lives much like the bees, like the subfamily Marsarinae.
Marsarinae are in the same family as wasps, Vespidae, and carry pollen in their honey stomachs. They share the wasp waist structure and are hairless, however DNA sequencing lists them as being closer to their predatory Vespine cousins than bees (Gess, 1996). The appearance of a wasp waist is not enough to determine if this is a bee or a Hymenopteran. The primitive sawflies lack this appearance (they appear in two parts: the head and thorax-abdomen fusion); it’s why we turn to molecular and fossil evidence to identify ancestry.
From Wasp to Bee: The Ammoplanidae Family
Ammoplanidae are one of the four families in the order Apoidea (the others being Crabronidae, Heterogynaidae & Sphecidae “sensu stricto” [I’ve included this classification term for the nerds]). Some members of the Ammoplanidae wasp family hunt thrips, a group of small insect that feed on pollen (Kirk , 1985). From this factor, Sann et al., suggest that the transition of wasp to bee came from Ammoplanidae’s predation on these thrips (Sann et al., 2018). Consider that if a wasp killed one during its meal, it would also consume the undigested pollen. Over time and random mutations, an Ammoplanidae could have developed a taste for pollen alone. When combined with nectar consumption, it makes for a complete vegetarian diet.
Bees First Emerged in Western Gondwana – Africa & South America
Charles Michner’s “Biogeography of the bees” lists Western Gondwana as the birthplace of bees*. Over 40 years later, Almeida and colleagues used data from the expanded fossil record to support Michner’s theory (Almeida et al., 2023). The environment was said to be dry and low in moisture, apparent by how bees today have a strong preference for such conditions along with seasonal temperate climates. For further proof, consider that you see the most bees in temperate areas, e.g. California or east coast New South Wales & Victoria.
When South America & Africa split 100 mya, the long-tongued bees from Apidae and Megachilidae (ancestors of the honey bees & leafcutter bees, amongst many others) split into African and South American lineages. The stingless bees stayed in Africa, the authors suggesting they went extinct in South America. By the time of the late Cretaceous 80 mya, all surviving bee families diverged into more species. The Colletidae family (plasterer bees) were the first to reach Australia, with the remaining families arriving much later (Almeida et al., 2023).
In the early Paleogene, 60-40 mya, intense weather shifts led to tropical conditions on multiple continents. In the Northern Hemisphere, the “boreotropical hypothesis” allowed for the expansion of the stingless bees, and other be lineages restricted to warm conditions. This enabled northward expansion from the equator, and perhaps drove adaptive mutations to these cooler environments. The migrations continued into the Neogene, 20 mya and the bees reached southern Asia.
*Land bridges, tectonic drifts and ice ages are beyond the scope of this article, and perhaps even this blog. The interested reader is referred to Almeida & colleagues’ work in full, and perhaps a textbook on earth sciences.
From Solitary to Social
This could be its own article
Communal nesting (“gregariousness”) means that a group of females share a nest where they lay their eggs and provide food to the hatchlings. Nothing more. No brood care or further feeding, nor the establishment of a hierarchy as to who forages and who tends to brood. Species within all bee families show communal nesting, from the mining bees to leafcutter & sweat bees. This contrasts the social bees, where the queen is the sole reproductive and the sterile workers tend to everything else.
Is it in these nesting communities that the jump from solitary to social was made? We could hypothesize that a random occurrence of intraspecific communication or changes in nesting biology drove these changes. “Intraspecific communication” refers to those styles used in finding and recognising potential mates, be it through pheromones, cuticular hydrocarbons or sounds.
Species that nest together must exchange information on matters like laying status, food availability and reproductive dominance. Leonhardt and colleagues suggest these signals are passed on by cuticular hydrocarbon profiles and ovarian activity (Leonhardt et al., 2016). The presence and absence of the respective compounds informs both the competitors and “helpers” about the reproductive potential of each female. The most mighty can use this to her advantage and induce helping behaviour in her subordinates. As nest size increased over time, the dominance signals and division of labour become more apparent, giving rise to higher forms of social nesting.
Why Pollen and Nectar over Animal Protein?
The evolutionary shift to pollen consumption started with the onset of flowering plants (the angiosperms). In combination with the forces that drive natural selection, availability, and plain random luck (“This flower tastes good, I want more”), bees eventually became a successful group of pollinators in today’s ecosystem.
Pollen is plentiful and reliable; you can track its availability by the season. It does not require the insect exposing itself to the risk of hunting and killing, since prey fight back. Some flowers provided both nectar and pollen, making it more desirable as the bees to collect a complete macronutrient profile (protein fats and carbs) from a single visit.
The niche of specialising in pollen, and subsequent specialisation within these niches reduced the competition for food resources. Plants took notice of course, and some evolved to only be pollinated by certain species, like the deep anthers and nectaries of Darwin’s Orchids from Madagascar. As bees progressed down the vegetarian path, their bodies developed branched hairs, and pollen baskets on their hind legs for better foraging. Others like the bumble bees, went down the path of buzz pollination, vibrating their bodies to release pollen from select flowers.
These are some of the factors that made pollen more desirable than animal protein. If I was to rank them by order of importance, it would be:
- Less competition
- No hunting prey
- Ability to be stored for extended periods (nectar as honey, pollen as bee bread)
- A nutritionally balanced diet
Doesn’t it all Sound very Wishy-Washy Hocus-Pocus?
You’re right. Beekeepers do not need a formal understanding of evolutionary biology to be successful. In fact many are successful because they spent more time inside hives than in books. To many beekeepers, a lot of this sounds like fiction… You mean to tell me that these invisible threads called “DNA” are compared using complex models and software to draw conclusions about lifeforms millions of years old? I don’t blame you.
These days, a lot of scientific papers are retracted for overexaggeration and falsehood. “How far can the scientific evidence be trusted?” This was the spirit of reader John’s reply. He has a point. I wager the number of readers who understand (or god forbid, enjoy) these advanced concepts are fewer than those readers who want to read about good hive insulation or how to requeen colony.
Read the whole post and found puns, typos mistakes? Voice your displeasure with a strongly-worded message to admin@tgbee.com.au, or by filling this form..