Snowflies amputate their legs to avoid freezing

October 4, 2023 • 12:10 pm

This paper recently came out in Current Biology, and I’m not sure you can access it by clicking on the title below or on the link to the pdf here, but I think you can, particularly if you have the (legal) Unpaywall app. But I can give all the relevant details in this post.

The main message is that four species of snowflies (a group of wingless craneflies in the genus Chionea) have developed the ability to tolerate temperatures that would kill nearly every other insect, both though unknown physiological adaptations as well as through a behavior that made this paper get noticed in the press: they amputate their legs when it gets too cold. This was observed in both the field and lab (well, no real amputation was seen in the field, but there were lots of legless snowflies found), and seems to be adaptive because the amputations prevent ice that forms in the legs from spreading to the rest of the body, which is fatal. Although other insects and even some vertebrates self-amputate limbs (it’s called “autotomy”), they do so in response to pulling, as when they’re caught by a predator.  These four species are the first ones known to self-amputate limbs in response to cold stress.

Here’s one of the species they studied, Chionea alexandriana (photo by Bryan Kelly-McArthur). Yes, it’s a fly, but its ancestors lost their wings. They don’t eat as adults, so they don’t live long.

Click to read or try the pdf link above:

Snowflies are found in boreal and alpine areas of the northern hemisphere, and are observed to be active in the snow at subzero temperatures.  Why do they want to do this? The authors speculate that it’s to take advantage of a lack of predators and also to find mates, which is easier when everybody’s on the white snow.  Because of this advantage, evolution has equipped them not only with unstudied physiological advantages that keep them moving when other insects die, but with the ability to self-amputate their legs when the legs begin freezing.

First, the authors did lab studies to see what low temperatures wouldn’t kill these species, compared to their sister group of craneflies. Here are the data, clearly showing that many snowflies can still get around when the cold-plate temperature (a bit lower than their body temperature) are as low as -9ºC, which is about 16ºC,

This is, as the authors said, is probably adaptive in helping the insects avoid predators (which are largely absent in winter), but also find mates. Here’s a funny paragraph from the paper:

We and others have observed snow flies audaciously mating in full view on the surface of the snow for 30 min or more.

Audaciously!

But on to the wings. The authors enlisted “citizen scientists” to collect snowflies from the snow, and they came up with 256 of them.  20% of these were missing legs when found, suggesting some mechanism for leg loss in the wild. It could be predation, but there are few predators. Some snowflies in the wild were observed crawling about with only three legs!

The authors found the solution when seeing what the snowflies did on cold plates in the lab. Sure enough, when the temperature got low enough that ice crystals began forming in the legs, they’d amputate the legs within half a second, a period long enough to prevent those ice crystals from ramifying through the body, which invariably kills the insect. (You might know that ice crystals can’t form except around a small nucleus of other stuff, and once that happens the crystals grow rapidly.)  Thus amputating the leg keeps the insect alive.

Cold plate measurements substantiates that.  Flies that amputated a leg in the cold lived 77 seconds longer than those that didn’t, which may allow them time to burrow below the snow, where the temperature is higher than the air temperature. Self-amputees also showed frozen bodies at significantly lower temperatures than flies which didn’t amputate their legs.

The ultimate test, of course, would be to observe differential survival in the wild, i.e., in the snow, but that would be hard to do, as you can’t control for temperature out there. However, given the habitat of these insects, and their differential survival in the lab when they undergo autotomy, the authors’ hypothesis seems reasonable.

Here are some videos of the snowflies amputating their legs in the lab. Trigger warning: self mutilation!

A para from the paper:

The key difference between leg amputation in snow flies and other insects appears to be the triggering stimulus. Many insects, including other crane flies, self-amputate legs in response to mechanical stimuli, such as pulling on the leg. The receptors responsible for sensing mechanical stimuli and triggering leg amputation in other insects are likely campaniform sensilla. However, we found that mechanical manipulations never triggered leg amputation in snow flies (Figure 4A). We hypothesize that leg self-amputation in snow flies may instead be triggered by thermosensory neurons that detect the temperature increase following ice crystallization of the hemolymph. The rate of ice propagation (0.5 s from the tarsus to the tibia) would provide ample time to execute this leg amputation reflex.

The authors detected leg freezing because when ice forms (an exothermic process), the temperature of the leg increases briefly. That’s how they were able to show how long it takes to amputate a limb when it freezes.

Here’s a video of a snowfly moving around in the field with only three legs. OY! Note: they cannot regrow their legs.

Now there’s a bit of the paper where the authors show exactly where the legs are self-amputated (the same place in all flies, and also in craneflies that self-amputate in response to predation), but that’s not all that absorbing.  Let’s just learn the lesson above, but also learn about those animals that undergo autotomy in response to stimuli other than cold. From the paper:

We found that snow flies rapidly self-amputate freezing limbs to prevent ice from spreading to the rest of the body. To our knowledge, amputation has not been previously described as a mechanism to avoid freezing and prolong survival at cold temperatures. Limb self-amputation occurs in many species, including reptiles, amphibians, mammals, birds, fish, echinoderms, crustaceans, spiders, and insects. It is typically used to avoid capture by predators, reduce cost of injury to a limb, escape non-predatory entrapment, or survive complications during molting. In snow flies and other crane flies, self-amputation consistently occurs at the joint between the trochanter and femur (Figure 4B). Past researchers have also collected snow flies from the wild with legs broken off at the trochanter. Specialized muscles that control amputation at this joint have been described in stick insects and crickets. Based on similarities in the breakage plane  and snow fly leg musculature, snow flies appear to use a similar amputation mechanism.

33 thoughts on “Snowflies amputate their legs to avoid freezing”

  1. “They don’t eat as adults”

    … 

    [ reads again ]

    “They don’t eat as adults”

    … I’m letting this sink in….

    I’m pretty sure this is one of those facts that I never would have predicted in a million years.

    1. Meaning that ANY animal could simply not eat.

      .. and now I’m wondering what that even means…

    2. There are lots of short-lived insects that don’t eat when they reach the adult stage. 

    4. Well I’ll be… that’s a personal best for a new level of understanding.

      Amazing, I’ll remember this the rest of my life, thanks.

      1. To put it into context, the adult stage of insect life cycles that doesn’t eat is generally pretty brief and wholly devoted to finding a mate and reproducing and it follows a usually rather longer larval stage that is pretty much solely* devoted to eating (it also has to avoid being eaten).

  3. Maybe we need to make tiny artificial legs for them. I know that Muskrats when trapped will chew off the leg to escape.

    1. I was about to ask which mammals practiced self-amputation but your answered my question. I will never hear “Muskrat Love” the same way ever again.

  4. Well, cut off my legs and call me shorty! As for winglessness, there are other insects that are active on the snow, and they too can be wingless. Flying takes a lot of energy, so cold insects won’t be big on flying anyway 

    1. Yes, a high proportion of moth species that have the adult life stage during the winter months have females that are wingless (or have very small vestigial wings that are incapable of flight). Their energy reserves are devoted to egg production. The males of these species are winged and fly on milder nights in search of females.

  5. ice crystals can’t form except around a small nucleus of other stuff, and once that happens the crystals grow rapidly.

    This isn’t true below about -36C = -33F. Also, freezing can be started by vibrations at higher temperatures. http://www.cas.manchester.ac.uk/resactivities/cloudphysics/topics/formation/

    In most situations, a little bit of dust, impurity, or even little vibrations in the water provide nucleation centers for the water to freeze onto. But if your water is very pure and very still, there is nothing for the water molecules to crystallize onto. As a result, you can cool very pure water well below zero degrees Celsius without it freezing.

    https://www.wtamu.edu/~cbaird/sq/2013/12/09/can-water-stay-liquid-below-zero-degrees-celsius

  8. Amazing. Thank you for posting such an interesting story, Jerry. I enjoyed reading it. 

      1. I can’t remember if Dawkins discusses this question in “Climbing Mount Improbable”, but the ability to load the blood (body cavity fluid, e.g. “hæmolymph”)with (relatively) small molecules that reduce the freezing point is distributed quite widely phylogenetically. I very much doubt that it would be through a common origin, buy multiple “discoveries” by different genera whose environmental preferences exposed them to conditions of occasional or intermittent freezing.
        It seems to me a good case of an “adaptive landscape” where a taxon can find a local “fitness peak”, start to adapt to optimise that characteristic, and so isolate themselves from other, possibly better, “fitness peaks” previously in their adaptive “space”.
        Damn, I’m suspecting now that that is an example from Dawkins, and I’ve just internalised it. Have to add re-reading “Improbable” to the to-do list.

    1. You never saw a one-legged punk pogo-ing?

      That’s a point – since the telly is currently full of some bondage-dance competition cross-over called “Strictly”, which I’ve never seen, I assume they have have “wheelchair ballroom” and “swimming pool tetraplegic samba” as categories. Maybe I should ask Auntie – the response from their production team could be … interesting. Interesting in a “Rick’n’Morty sense of “strained”.

  12. Pretty darn cool I tells ya. Thanks for the post. It’s hard to find a niche that isn’t filled with some form of life (and I mean larger than microscopic life). Though the Antarctic is devoid of insects iirc.

  13. An obvious question—at least to me—is whether the flies actually mate after dropping their legs, and how. (Good manners prevent me from even trying to imagine two legless flies rolling around in a mating embrace.) While self-amputation might permit the flies to live longer lives, it’s hard to believe that a fly’s chance at mating actually improves in such a condition. Maybe they’re OK if they amputate only one, or two, of the six legs they had originally. Go figure. And how, after their legless romp, do the females lay their legs, and where? Do they just drop them on the snow? Without legs, where else could they go? So many questions—most of which make me chuckle.
    🙂

    It’s interesting to see that the flies retain their halteres, even though the have lost their flight wings.

    Thank you for the interesting and fun article.

    1. A phrase from my undergraduate textbook on mollusc palaeontology sticks in my mind. Being a teenager with the American Bomb in my backyard, “copulation by guided missile” is … evocative.
      The behaviours may not fossilize, but the associated anatomy (sperm-carrying extra-long, detachable tentacles ; calcite darts with spermatophore cavities) do fossilize.

  14. “… when the cold-plate temperature (a bit lower than their body temperature) are as low as -9ºC, which is about 16ºC, …” – that should be 16ºF, I believe.

    Also, which vertebrates autotomize their limbs? Tails, yes, but I’ve never heard of a vertebrate voluntarily dropping a limb.

Comments are closed.