Mantis fly has a walking pupa that climbs trees before hatching

December 6, 2017 • 12:45 pm

Although this observation and video don’t revolutionize science, they do provide some information that was missing about the life cycle of an insect, and, importantly, they allow the lucky finder to see something that presumably no other human had seen before.

In this case it’s a substantial part of the lifecycle of a “mantidfly“, Ditaxis biseriata. Mantidflies aren’t “true” flies, which are in the order Diptera, but fit within the order Neuroptera along with lacewings, antlions, and various relatives.  Here’s the species at issue, and you can see from its forelegs why it’s called a “mantidfly”. It’s obviously a predator, and eats other insects:

Ditaxis biseriata, adult

Within the order Neuroptera, this species is in the superfamily Mantispoidea and the family Mantispidae. And in that family there are four subfamilies. The immature life stages of one of them, the Drepanicinae (of which D. biseriata is a member) wasn’t known or seen until now. The German Wikipedia entry (the only one) says this: “Die adulten Tiere leben räuberisch von anderen Insekten, über die Präimaginalstadien ist bisher nichts bekannt”, which I take translate as “the adult insects are predators on other insects, but the pre-imaginal stages aren’t yet known.”

But now they are. The discovery is reported a new paper in Biodiversity Data Journal by James B. Dorey and David J. Merritt (reference and free link below) and summarized in EurekAlert!   The latter article says that Dorey is an entomology student at the University of Queensland, who saw a mass emergence of these creatures in his father’s macadamia nut orchard in northern New South Wales. Not knowing what they were, Dorey took a video and obtained some eggs from a female that produced larvae that could be identified. His mentor and lecturer, Merritt, helped identity them as D. bitaxa.

So the life cycle is this: the eggs are laid by mated females who live aboveground, which are presumably laid on trees and then hatch, with the larvae heading underground. What they do under there, how long they’re underground (remember, cicadas can live underground as preadults for 17 years), and what they eat, is still unknown: the relatives in the family eat spider eggs as larvae, but we don’t know about this family. The larvae probably molt several times, producing successively larger individuals (the stages are called “instars”), and then at some stage they form a pupa.

The remarkable thing about the pupa, unlike a butterfly cocoon or a Drosophila pupa, is that they’re mobile: the late pupa, which contains an adult insect that hasn’t yet hatched, is called a “pharate”, and in the case of D. biseriata it can move!  This freaks me out, but here’s the video that shows a pupa who has already climbed a tree hatching (“eclosing”) into an adult. This is fricking amazing, at least to someone like me who studies insects that have immobile pupae. The pharate you see that hatches into an adult has walked about two meters up the tree, become static for a while, and then eclosedfrom its pupal case. This, taken in the wild, is a remarkable video as well as a first observation.

Be sure to watch this enlarged for full wonderment!

Here are some pictures from the paper. First, a pharate that’s walked up the tree and is ready to hatch:

And the eggs and then the larvae that go underground (caption from paper):

Ditaxis biseriata eggs and larvae. Top Left. Egg batch before eclosion; Top Right: egg cluster after eclosion of first instars, seen clustered on the egg batch; Middle: dorsal view of preserved first instar; Bottom: dorsal view of live first instar during locomotion.

So that’s what’s new, besides the fact that mass eclosion of the type seen by Dorey is also new: this phenomenon has never been reported in the family.

h/t: Matthew Cobb


Dorey JB, Merritt DJ (2017) First observations on the life cycle and mass eclosion events in a mantis fly (Family Mantispidae) in the subfamily Drepanicinae. Biodiversity Data Journal 5: e21206.

10 thoughts on “Mantis fly has a walking pupa that climbs trees before hatching

  1. Very cool. I worked on a gall-forming fly (a true fly), the pupae of which move, burrowing to the edge of the gall. But this is something else.

  2. I’m don’t understand the difference between this metamorphosis and the way the cicada do. Cicadas live underground and then at a certain time crawl up tree trunks and climb out of their skins as adults. What is the extra significance here?

    1. This is also a metamorphosis, but in the case of the cicada it is classified as an ‘incomplete metamorphosis’ because the change is less dramatic The juvenile cicada is a nymph, with an anatomy that is not greatly different from the adult.
      Mantidflies have a more dramatic metamorphosis from larva to pupa to adult. It is not unique that the pupa is mobile. But the list of mobile pupae is not long.

  3. This is very cool, as mantidflies are cool (way cool!). But walking pupae are also seen in the Corydalidae (Dobsonflies).
    The larva of the above species of mantidfly might be a predator of soil arthropods, as is known for other species in the group.

  4. Manatispids are not at all common where I’ve lived in the Pacific NOrthwest. I’ve seen two individual adults, 2 different species. One in the Siskiyou Mountains in sw Oregon, the other at my boyhood home in the Nooksack Valley of nw Washington. Both just landed on my arm. Beautiful and unforgettable.

    Pharate pupae may also be decticous, with articulated, functioning jaws. Happens in my spirit animal, Caurinus dectes [Mecoptera, family Boreidae].

  5. Fascinating. If my observations are correct, the mosquito pupa is likewise mobile; it looks a bit like a football or oval with a coupla antennae poking out the one end, and they’re wicked fast swimmers. The larva is the well-known wriggler; the pupa is somewhat smaller, more compact, and way faster.

Leave a Reply