Getting to know flower wasps
Flower wasps have been teasing me for weeks. I’m seeing them every day. Many species, varying in size and colour. Their behaviour is fascinating, their biology mysterious. And I don’t even know their names.
I decided it was time to investigate.
After reading all the papers I could find on Thynninae, I better appreciate the challenge of flower wasp species ID. I also have a better understanding of the taxonomic position of the group and how to recognise a flower wasp when I see one. Importantly, I’ve also read just about all there is to read on flower wasp life cycles and behaviour. In a nutshell … there’s not much. This all reassures me that any observations we make are valuable – even if we can’t put genus or species names to any of them. Yet.
There was a lot to think about and learn … so this is quite a saga. There are three sections:
The taxonomy: sorting out who’s who
The natural history: what is known, what I’ve been seeing, and what more I should look for
Identification: how to recognise flower wasps (and other wasps too!)
The taxonomy: sorting out who’s who
Our usual approach to nature study begins with species identification. We seek to work out who’s who so we can read up on that genus or species. What we learn from published research then informs our own observations. I’m a firm believer in ‘the more you know, the more you see’.
But this approach doesn’t work well for flower wasps. Even when I have good photos, from a variety of angles, showing both sexes, my identification gets stuck at the subfamily – Thynninae.
After just a couple of days reading I realised why I’ve been finding this group so challenging.
There are rather a lot of flower wasps and many have yet to be formally described and named. A leading researcher in the field, Graham R. Brown, estimates that in addition to around 600 named flower wasps there are more than 2000 undescribed species and many unrecognised genera (ref. 4, 5).
Of course, flower wasps are not the only insects to which this applies. But for flower wasps there aren’t many options for getting to first base with identification – not without the insect under a microscope, and perhaps a trip to a museum.
One of the techniques we normally start with is simple photo-matching. Can we find a good match with images of a named species? There are plenty of photos of named flower wasps online … and most have been named by photo matching. You can see the problem. Many of the identifications are wrong.
It’s not surprising that there are so many errors. Flower wasp genus names have changed rather a lot. Such taxonomic revisions are important, of course, but they do create chaos in online image repositories, including the Atlas of Living Australia.
Even the family name is a trap. Australian flower wasps belong to the subfamily Thynninae. Until recently, Thynninae was a subfamily of Tiphiidae. However, Thynninae has been moved into a new family, Thynnidae, on the basis of molecular evidence (ref. 8). There is still a family called Tiphiidae, but it now includes only one Australian species. Confused? Rest assured you are not alone. Even major data systems are at odds. Neither the Australian Faunal Directory nor ALA reflect the change, but iNaturalist does.
In addition to the confusion around taxonomies, species are always going to be difficult to identify from field photos alone. The distinguishing features are often very small or hidden structures, such as genitalia.
I now accept that most of our flower wasp sightings will simply be labelled ‘Thynninae’. For now, anyway. And that’s OK. Anonymous wasps are still deserving of study.
The natural history: what is known and what I should look for
I’m convinced that it’s worth watching, photographing and documenting flower wasps … even if I never get to know their names. The information may be useful to someone, somewhere, sometime. And anyway, wasps are fascinating and more than a little bit photogenic.
What’s already known?
Flower wasps have gained notoriety for their role in orchid pollination (e.g. this Australian Geographic story or this in The Conversation ). Yet despite such fame, the natural history of flower wasps remains poorly documented (ref. 7).
Most of what is known comes from a couple of well-studied Australian species by Ridsdill Smith (ref. 9, 10). In summary:
Adults emerge and mate. Each Spring, adult wasps emerge from the soil. The males fly fast and low, seeking females. The females emit chemical attractants, even before they leave the soil. They then emerge and, if not immediately grabbed by a waiting male, will perch on low vegetation and continue to emit pheromones.
Males carry the flightless females away from the emergence site before courtship and mating. After considerable mandible stroking, the pair couples and the female stops releasing chemical cues.
Connecting complex genitalia is no simple task. And it’s not surprising that the apparatus is so complex. As well as the usual function of insemination, the structure needs to be secure enough to hold the female for an extended period … and it needs to pivot.
(It can take time to achieve coupling, and it doesn’t always work. This pair were interrupted by another male before they managed to connect)
Feeding. Females rely on males for food. In most species the male feeds his mate a sugary liquid from his mouthparts or from a storage area just beneath his head. In some species he carries her to flowers or other food sources to feed. Either way, without this coupling the female cannot access a meal.
Egg-laying. The male tends to return the female, mated and fed, to near where he found her. She will usually dig into the soil immediately, or crawl just a short distance first. She searches underground for a large scarab grub, using her mandibles and legs to dig. Once she has found a suitable grub she paralyses it with a well-directed sting directly into the nervous system. She then drags the grub deeper into the soil, makes a chamber, and massages the inert (but living) body until it is soft and relaxed. She then lays a single egg … and heads back to the surface. She repeats the mating-feeding-egg laying cycle about 3 times, over a week or two.
Larval development. At first the wasp larva feeds on the grub’s body fluids through a small hole in its body wall. Eventually the host dies and the now much larger wasp larva consumes the grub’s internal organs. The wasp then pupates and eventually emerges from the chamber as a fully-developed adult.
What’s left to learn?
So that’s the story for a few species, but what about the many, many others? With such an array of Thynninae here in the forest, there’s plenty for me to observe, document and ponder. For example:
What food sources do adults use?
Male flower wasps are quite selective in the flowers they visit (ref. 7). They most commonly choose Myrtaceae including Eucalyptus, Angophora, Leptospermum, and Melaleuca. There are also records of them visiting Hakea and some forbs such as Scaevola and Pimelea. Notably for us, they also use Xanthorrhoea – at least in W.A! There is some evidence that they also feed on insect secretions – honeydew.
(In 2018 we spent time in the south-west of Western Australia. Orchids, rare birds, extraordinary vegetation … and flower wasps! As reported by Brown and others, the males were indeed feeding on Xanthorrhoea and Eucalyptus.)
Surprisingly, I’ve never seen our home wasps feeding at flowers – but perhaps that just reflects the height of the trees. The Angophora and Eucalyptus flowers are always way beyond my reach, even armed with a long focal length camera lens.
But this year the trees are not a feeding option for our wasps anyway. The Fire killed the Hakea, and set back all the other trees and shrubs such that they’ll not flower this season. This begs the question of just what they are using. I know they are feeding, as I’ve seen males feeding females. But where are the males getting the sugars?
One possibility is the psyllids and other sap-sucking bugs. There are plenty of lerps secreting honeydew on the epicormic growth. I also plan to closely monitor our grass trees – they are starting to flower, and there are hundreds of spikes! A third possibility is the white Pimelea flowers. They’re flowering profusely, so they warrant some close monitoring too.
How do the females feed?
Females are known to feed from regurgitated liquid from the male’s mouthparts or from a storage region beneath his head. Yet I think I’ve seen something else again … a feed droplet beneath the male’s abdomen.
I watched this pair for five minutes. During that time he twice flew to a different perch, but the female remained attached, curled over, with her head in the liquid droplet. The perches were always vertical plant stems of leaves, and always about 40cm above the ground.
I need to keep a look out for this behaviour. This is not something I’ve seen documented elsewhere, so it could be an interesting find.
What are the host species?
The species studied by Ridsdell parasitise a range of scarab larvae, and this is often extended to describe the life cycle of flower wasps generally. However, there are reasons to believe that, as a group, flower wasps target a much wider range of soil-living insect larvae (ref. 3).
This is clearly a difficult question to answer. I’m probably not going to be able to shed much light on the subject, but you never know. Even our growing biodiversity home lists say something about the potential range of hosts. And maybe one day I’ll be rearing a soil grub only to have a flower wasp emerge. Stranger things have happened at our house! Remember the Wonboyn Wasp story?
Do they miscouple?
Flower wasp couplets are not always a matched pair. This further confounds flower wasp taxonomy … females in museums have been labelled on the basis of the male they were found attached to, but sometimes he’s a completely different species (ref. 1, 6)! This raises questions about the female’s role in mate selection. In the absence of conspecific males, females may deliberately attract and accept males of other species simply as a means of feeding and transport.
I have seen what looks to be male indecision. Here the male flies a female to a leaf but departs without mating … although he seems unsure.
On another occasion, one female, two males (or perhaps 3) … and the males appear to be two different species.
Do I ever see apparently identical males with very different looking females? I need to keep watching and comparing my photos. If so, it’s interesting, but not really evidence of miscoupling given that I can’t confirm the male’s identity. An interested person would need to collect such pairs for a closer look. Perhaps, someone will, one day … in the meantime, I’ll just keep recording what I see.
Identification: how to recognise flower wasps
On the one hand, recognising flower wasps is simple. Two wasps joined end-to-end, with one partner smaller and wingless … flower wasps. Can’t be much else.
Mating pair of flower wasps – Thynninae (3/9/2020)
Similarly, a wingless insect with strong mandibles, no obvious petiole, no ocelli (usually), perched motionless and head up on low vegetation … odds are it is a female Thynninae.
Wingless female - Thynninae (11/11/19, Nadgee NR)
But how can I confidently recognise a single male flower wasp when I see one? There are many other large wasps, most notably the spider wasps (Pompilidae), potter and paper wasps (Vespidae), ‘hairy’ flower wasps (Scoliidae), diggers and mud-daubers (Crabronidae, Sphecidae).
Some identities are obvious, some less so. The ‘it just looks like one’ strategy often works, but it’s not always ideal. Objective criteria are better.
I decided to create cheat sheets for distinguishing various wasp families. A list of features to help rule groups in or out. And, ideally, based on structures that can be seen in a photo. So, using the seminal text on Australia’s insects, CSIRO’s Insects of Australia (ref 11), I’ve put together the following. Not comprehensive, not perfect, but it’s a start.
Flower wasps are best defined by what they’re not
When it comes to recognising Thynninae, it’s easiest to first rule out the other main contenders.
Thynninae eyes aren’t strongly indented
If the eyes have a large indent in the inner margin, it’s probably Vespidae, Scoliidae or perhaps Crabronidae (Trypoxylini).
Thynninae forewings aren’t folded
Folded wings are another feature of Vespidae. Thynnidae don’t do this, nor do many other groups
Thynninae wings don’t have a tight array of ridges
Scolidae, the so-called ‘hairy flower wasps’, have very characteristic wings.
Thynninae legs aren’t particularly long
Long, slender legs with strong spines …. Pompilidae. Or perhaps Sphecidae.
Thynninae do not have an elongated petiole
The second metasomal segment is particularly elongate in Sphecidae. The waist can also appear long and thread-like in some species of various other families … but not in Thynninae.
Thynninae don’t have a wide or square head
Crabronidae would not usually be confused with Thynninae. They are smaller and bee-like. Still, head shape is another criterion to cross-check against.
There are a few features that they do have
Check for a particular, small wing vein
Most Thynninae have a small extra vein crossing the first submarginal cell of the forewing, the ‘spur vein’. It’s surprisingly easy to see, once you know what to look for.
The pronotum in Thynninae is long(ish), dorsally
The ‘shoulders’ of Thynninae seem to wrap across the back. The shape of the pronotum, both dorsally and laterally, is quite helpful in distinguishing wasp families. The diagramatic cheat sheet above illustrates the differences. Thynninae and Pompilidae, however, are similar in their pronotum shape.
Thynninae have a ‘brow ridge’ – Anthoboscinae don’t
The antennae in Thynninae project from beneath a ridge. This ridge is often small and can be difficult to see – particularly if the wasp is hairy. The brow ridge feature distinguishes Thynninae from the other main sub-family within Thynnidae, the Anthoboscinae.
Behaviour can help too … sometimes
The list above makes the task of identfication seem simple. It often isn’t, at least not from photos alone. Wasps don’t always oblige by posing in quite the right way.
Fortunately, behaviour can provide important clues. A flower wasp is not going to be dragging a spider or building a mud nest. It’s not going to be digging a sand burrow or living in a papery nest with a bunch of other wasps.
But even behaviour can be misleading.
Last year I watched this wasp fossicking around under the grass. I confidently called it a spider wasp and assumed it was a female on the hunt for wolf spiders. Now I look at the photos with new eyes. It’s actually a male flower wasp, probably homing in on a concealed female. If I’d realised at the time I might have waited to witness the catch!
Beware the imposters
Finally, not every insect that looks like a wasp is actually a wasp. Always a good idea to look twice before jumping to conclusions.
References
Brown, G.R. 1993. A new species of Lestricothynnus Turner with notes on miscoupling in Thynninae (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society, 32: 197-99 open access
Brown, G.R. 1998a. The generic status of Catocheilus Guerin and Hemithynnus Ashmead (Hymenoptera: Tiphidae: Thynnini). General and Applied Entomology: The Journal of the Entomological Society of New South Wales, Vol. 28, 1998: [89]-92
Brown, G.R. 1998b. Revision of the Neozeleboria cryptoides species group of thynnine wasps (Hymenoptera: Tiphiidae): Pollinators of native orchids. Australian Journal of Entomology, 37: 193-205
Brown, G.R. 2000. Some problems with Australian Tiphiid wasps, with special reference to coupling mechanisms, pp 210-217 in Hymenoptera: Evolution, Biodiversity and Biological Control. Ed. Austin, A.D. & Dowton, M. (Editors). CSIRO publishing, Collingwood
Brown, G.R. 2005. A revision of Tachyphron Brown and description of two new genera within the Ariphron group (Hymenoptera: Tiphiidae). Journal of Natural History, 39(2): 197-239
Brown, G. 2015. Rugosothynnus gen. nov. (Hymenoptera: Tiphiidae: Thynninae: Rhagigasterini), a newly recognised Australian genus. Zootaxa 3925 (3): 361-386 online
Brown, G.R. & Phillips, R.D. 2014. A review of the diet of flower wasps. Northern Territory Naturalist, 25: 50-63
Pilgrim, E.M., von Dohlen, C.D. & Pitts, J.P. 2008. Molecular phylogenetics of Vespoidea indicate paraphyly of the superfamily and novel relationships of its component families and subfamilies. Zoologica Scripta, Sept. 2008. Vol. 37(5): 539-560
Ridsdill Smith, T.J. 1970a. The biology of Hemithynnus hyalinatus (Hymenoptera: Tiphiiidae), a parasite of scarabeid larvae. Journal of the Australian Entomological Society. 1970. 9(3): 183-195 open access
Ridsdill Smith, T.J. 1970b. The behaviour of Hemithynnus hyalinatus (Hymenoptera: Tiphiiidae), with notes on some other Thynninae. Journal of the Australian Entomological Society. 1970. 9(3): 196-208 open access
Naumann, I.D. 1991. Hymenoptera (wasps, bees, ants, sawflies). Chapter 42, pp916–1000, in Insects of Australia, Second Edition, 1991. CSIRO (Division of Entomology). Melbourne University Press, Carlton.
Most of the photos were taken here in the forest. Those taken elsewhere are labelled with the location.
The cheat sheets are simplified descriptions and may include omissions and errors. If you have expertise in wasp ID, and you spot a glaring error, I’d be pleased to hear from you.