I do these occasional posts about science papers. Some are just for fun. But sometimes — honest! — there’s an underlying connection to the greater Crooked Timber project.
This post is one of that sort, because it’s about the limits of understanding. Unsurprisingly, it involves biology.
So we all learned back in high school that our nerves are sheathed in a coating, like insulation on a copper wire. The coating is made of a special substance called myelin. If your tenth grade biology text mentioned myelin, it probably said something like “myelin allows impulses to flow along the nerves faster and more efficiently”. Which is true! It may also do some other things, but “myelin = faster and more efficient transmission” is what we all learned in sophomore biology back when, and it’s basically correct.
Occasionally something goes wrong, and either the myelin sheath doesn’t form right, or the body’s immune system gets confused and attacks it. This can lead to serious problems, conditions like multiple sclerosis and muscular dystrophy. Also, newborn babies haven’t finished forming their myelin sheathing yet. That’s why newborns are so very weak and uncoordinated. That magic moment, around the three month mark, where the kid suddenly starts holding up their head, looking around, and intentionally reaching for stuff? That’s when “myelinization” is complete.
Myelin itself isn’t alive. It’s an “extracellular material” made of a complicated mixture of fats and proteins. But it’s not a passive, inert coating. It does stuff, and it’s manufactured and maintained by special cells that hang out around your nerves. You have hundreds of millions of these cells, and they spend a lot of time, effort and energy creating and maintaining your myelin. In fact, the whole system is metabolically quite expensive to build and then to maintain. So, your body conserves resources by putting more myelin on important nerves, and less or none on nerves that are peripheral or only used rarely. This includes some nerves for sensing pain.
You might think this is a bad idea: surely you want to respond to sudden pain quickly? Well, evolution has a hack for that: the “spinal reflex”. That’s where a subroutine hardwired into your spinal cord will yank your hand away from the hot stove before the pain has time to reach your brain. This is why, when you stub your toe, there’s the initial sensation of stubbing followed by the “AAHHH” wave of pain half a second later: that particular subsystem includes some unmyelinated nerves.
[first he bites, then he grinds]
Myelin plus its network of builder / maintainer cells together make up something like a third of the total mass of your nervous system. You’re probably carrying around a couple of kilos of the stuff, several pounds, and it’s burning a non-trivial proportion of your body’s energy budget. But if some Department of Metabolic Efficiency were to magically make your myelin disappear, well… it wouldn’t be great. You’d probably drop dead on the spot. Maybe you could survive on a ventilator, but you’d have severe muscular dystrophy forever. All that metabolic energy is being spent for a reason.
Okay so: For a long time after it was discovered, scientists used to think that myelin was unique to vertebrates.
Invertebrates have nerves, obviously. But most invertebrates are small. A housefly has much faster reflexes than you, but that’s not because its nerves are fast. They’re not fast. Insects don’t have myelin, so the fly’s nervous impulses travel actually travel more slowly than yours do. Quite a bit more slowly. But the fly’s entire body is just a centimeter long, so those impulses don’t have to travel very far. So the fly’s reflexes are a lot faster than yours. The fly may have a slow nervous system, but it can still avoid your slap.
And the fly doesn’t have to spend scarce metabolic resources on building and maintaining expensive myelin.
There’s some evolutionary history here. “Primitive” vertebrates like hagfish don’t have myelin. And here’s a thing: hagfish are scavengers, rather sluggish, and they don’t grow very big. So a slow nervous system isn’t a huge handicap for them.
There’s some evidence that early vertebrate fish may have evolved myelin back in the middle Paleozoic, allowing some fishes to suddenly get much bigger, much faster, or both. You know how there was a page in your childhood paleontology book that was just trilobites and whatnot, with a few small, simple jawless fish derping along in the background?
[copyright G. Paselik, Humboldt Museum]
and then you turned the page and, wham, there were big damn predatory fish with terrifying bear-trap jaws?
Current thinking is that’s the moment when early vertebrates developed myelin. Because without myelin, it’s really hard for a vertebrate to grow big. Your nervous impulses move too slowly for your large body to react to threats or prey. With myelin, though — well, it’s expensive stuff, but it lets vertebrates grow large (dogs, horses, humans) and sometimes ridiculously huge (elephants, dinosaurs, whales). And it also lets large animals coordinate large bodies and move quickly.
Okay, so vertebrates got myelin so they could grow big. And for decades this was the paradigm: vertebrates are large, so they have myelin. Invertebrates are small, so they don’t.
But then, problems arose.
1) There are large invertebrates. What about octopuses and lobsters? Heck, what about the giant squid?
2) There are small vertebrates. Shrews, newts, guppies… lots of little vertebrates are no larger than large insects.
3) And… it turns out a bunch of invertebrates /do/ have myelin. And the ones that do, aren’t necessarily large.
In short, it’s a hell of a mess. Let’s just tick a few points.
Octopi and squids
This one is relatively easy. If you want to make nerve impulses travel faster, but you don’t have myelin, there is an alternative: you can make the nerve axons longer, and you can make the nerves themselves thicker. You might remember that electrical resistance in a wire decreases as the cross-section of the wire gets bigger? It works that way with neurons too. And this is the option that octopi and squids have chosen. These large cephalopods have some extremely long axons — like, inches to feet long — and also some extremely thick nerves that are very large in cross-section.
Furthermore: large cephalopods seem to have relatively decentralized nervous systems. You and I have spinal reflexes. Well, an octopus takes that to the next level. Each individual arm has complex nerve ganglia that let it do all kinds of actions and reactions on its own, without the central nervous system getting involved. Presumably the brain can override when needed. But much of the time, the octopuses’ arms are autonomous systems. This dramatically cuts down on the need for long-distance nervous communication, and thus on the need for myelin.
Giant axons pop up in some other invertebrates too, most notably in crayfish — relatively large, fast-moving invertebrates that rely on quick reflexes both for predation and escape.
Okay, so that’s sorted. But now we jump right off the tracks.
Small vertebrates
We have no goddamn idea.
Seriously. A small guppy is the same size as a large spider or beetle. The guppy invests in myelin while the spider and beetle don’t. Why? The guppy is only a few centimeters long. The boost to its reaction time is going to be very modest, and it’s paying a metabolic price for this. It’s not giving the guppy an obvious advantage. At this scale, guppies and other small fish are regularly captured and eaten by water spiders, diving beetles, dragonfly larvae, and other predatory invertebrates. So why does the guppy bother making myelin?
We don’t know.
But then it gets worse.
Some invertebrates have myelin
This post is already running long, so I’ll keep this short: myelinization has evolved independently at least four times outside of the vertebrates. But it has evolved in invertebrates that aren’t very large and that don’t seem to have a compelling need for fast reflexes. The myelinized groups include several species of marine invertebrates, including copepods and tiger shrimp, and also a number of worms and wormlike creatures, including earthworms.
First off, here is the occasional paper. It dates from 2007, but the field hasn’t really advanced much since then. If you want a TLDR short version, this page is pretty good.
*Now when I first went down this particular rabbit hole, I hit a point where I stopped for a reality check. Was this stuff as bizarre as it appeared? Or was I just missing something obvious? Eventually this led to an e-mail correspondence with a very pleasant emeritus professor of invertebrate physiology. This distinguished senior academic took the time to politely explain to me that, yes, this whole thing was every bit as weird and mysterious as it seemed. It wasn’t an artifact of my poor research skills or limited understanding. The evolution of myelin is — they cheerfully reassured me — some authentic Deep Science WTF.)
Right, so. Some invertebrates have myelin, like tiger shrimp. You know tiger shrimp? Sure you do. Remember the last time you had dinner at an Asian fusion place?
Tiger shrimp are large-ish crustaceans that, by pure evolutionary accident, are delicious with soy sauce. Morphologically they’re pretty similar to crayfish. They’re both about the same size, and they’re both predatory clawed aquatic crustaceans.
But they’ve independently evolved completely different ways to speed their nerve impulses along. The crayfish uses giant axons, while the tiger shrimp uses myelin. Oh, and the tiger shrimp appears to hold the record for fastest nerve impulses in the animal kingdom. It runs at about 200 meters / second, which is over 400 mph or 600 kph. That’s several times faster than human nerve impulses. Since the shrimp is only about 15 cm long, this means a nerve impulse can go from its brain to its tail in a fraction of a millisecond. Why does the shrimp need literally lightning reflexes? We don’t know.
Keep in mind here that most crustaceans have neither large axons /nor/ myelin. Crabs don’t. Most shrimp don’t. Lobsters are predatory clawed aquatic crustaceans just like crayfish and tiger shrimp. But they’re much larger. A big Maine lobster can get up to almost a meter long and weigh over 8 kilos or 18 lbs. That’s the size of a large house cat. You might think an animal that size would invest in speeding up its nervous impulses. Nope. Lobsters don’t bother with either big axons nor with myelin either, and they seem to do just fine.
And then there are copepods. Copepods have separately evolved myelin. But copepods aren’t large! They’re tiny.
They’re nicknamed water fleas. They range in size from “smallish insect” to “barely visible speck”. They’re basically plankton.
Copepods might need good reflexes to avoid predators, but at that size scale, having myelin is going to make almost zero difference. So why are they investing in coating their teeny tiny little nerves with expensive myelin? There are thousands of species of invertebrate plankton animals that don’t bother with myelin. Why did copepods go to the trouble of evolving myelin, and why are they keeping it?
And then there are earthworms. Which… what the hell? What in god’s name does a blind, nearly brainless worm need with myelin? Earthworms aren’t large, they aren’t predatory, they don’t have very complex nervous systems. Yes, earthworms need to avoid predators —
[copyright 2025 False Knees — go check out False Knees, it’s great]
— earthworms need to avoid predators, but so do caterpillars and beetles and spiders and a million other kinds of small invertebrate, and they do it without myelin. Why are earthworms spending precious resources on an expensive tissue that they don’t seem to have any use for?
We don’t know.
So here’s the mystery: myelin has evolved a bunch of different times, in widely separate lineages all across the animal kingdom. But since it’s expensive to create and maintain, it must be serving some function. In the case of large vertebrates, that function is obvious: it lets us be large. And in the case of a few not-large invertebrates, you can maybe squint and say well… I guess that guy needs especially fast reflexes, because reasons? But that still leaves a bunch of creatures, several different sorts of invertebrates and small vertebrates, that are investing in myelin and we just don’t know why. Remember, most of these guys have close relatives or niche colleagues that don’t bother with myelin, nor large axons either.
Maybe myelin is doing something else, in addition to insulating your nerves? Biology loves to do that sort of thing. But what possible Something Else would unite earthworms, tiger shrimp, geckoes, copepods and guppies?
It’s a mystery. And since this is an area of study that sits awkwardly at the intersection of some unrelated sciences — neuroscience, cladistics, evolutionary biology, invertebrate physiology — it’s not attracting large amounts of research money at the moment. So it’ll probably remain a mystery for some time to come.
And that’s all.
{ 1 comment… read it below or add one }
Michael Cain 09.05.25 at 8:13 pm
Evolution is usually considered path dependent. Myelin strikes me as something that once you’ve evolved it, it’s hard to get rid of.