How exactly does a caterpillar morph into a butterfly?

Link here: How exactly does a caterpillar morph into a butterfly?

Excellent article, Doug, and easily understandable, in only four paragraphs (although the last one is a bit on the long side).

Why they can’t teach that in school, I have no idea.
One question: The imaginal disks: they store everything required for the resultant adult stage, including its brain?

I just wanted to say great job on the staff report Doug. You really got the entire concept explained very cleanly and well.

My daughter loves butterflies and has raised them for three years straight. I have never seen such a clear and concise explanation of the process.


That’s what I was wondering. The description in the report makes it sound like the brain is also reduced to goo and then reformed. I was wondering if the butterfly ended up with any memories from the time of being a caterpillar. Not that caterpillar memories would necessarily do a butterfly much good.

Does part of the brain stay intact through the process? Have there been any memory studies done?

The brain does not apparently stay intact through the process, but recent experiments (within the last few months, if I recall correctly) indicate that some memories are preserved. It’s still officially an open question, I gather, and a fairly hot one, given the implications.

Sorry to be a pedant but chitin is actually a long-chain polymer carbohydrate - not a protein, although it is similar to keratin.

Otherwise, fantastic reply to an interesting question.

Yes, thanks jehh87; Doug caught that, and it’s a correction that is in process of being made.

Staff Reports (like this one) are officially posted on Tuesday, but they’re available in advance (on Fridays), sometimes in draft-ish form.

Excellent job Doug. This process has bugged me since elementary school, including the brain and memory question. Considering that is over 40 years ago, it’s been a long time coming.

By the time I got to college my question morphed into “how in the heck did this evolve?” What is the evolutionary advantage?

And now after reading the article and being educated on imaginal disks, my new question is "do these imaginal disks contain DNA that is different than those of the caterpiller?

I don’t know why, but metamorphosis has always intrigued me.

Does the caterpillar/butterfly actually have an identifiable brain, as opposed to a system of distributed ganglia?

Insects have two groups of ganglia in the head, the supraesophageal ganglion and the subesopageal ganglion. The former is often referred to as the “brain”. There are more clusters running through the body, but the general trend of insect evolution is for the number to grow smaller, down to the single thoracic-abdominal ganglia in houseflies. (As a general rule, anything that has a head has a brain. The main point of a head is to put a centralized sensory complex there, and that needs a data processor.)

Yeah, but how does a caterpiller morph into a wasp? We sometimes raised them, waiting for the miracle and joy of butterflies, but once a wasp came out. My mother came close to doing the ceiling cling while I explained that parasite wasps sometimes get into caterpillars. It was one of the wonders of nature, but I had a hard time convincing my mother it was supposed to be wonderful.

Could be some sort of parasitic thing, like I’ve seen involved with ants, I think, on Animal Planet… I may be WAY misremembering this, but I think the wasp injects eggs into the host insect, and, well, THIS happens…

Anyone have the SD on this?


Here’s am interesting story.

A wasp? Parasitic wasps normally lay a hell of a lot more than one egg per victim.

Since this was the first question, I’ll start with this one.

As far as I’m aware, in the more “advanced” holometabolous insects (basically, all the insects with complete metamorphosis other than various Neuropteroids) nearly everything is “recycled” during the pupal stage, but I think a fair bit of the nervous system is retained. Yes, it undergoes some drastic changes in organization (especially the fusion of larval ganglia), but I believe a fair number of the cells - especially those in the head - and their connections to one another survive the transformation. It would not surprise me to find that the degree to which the nervous system is broken down varies from group to group, but I doubt that anything has 100% broken down.

The next question was about chitin, and yes, chitin is a polysaccharide like cellulose, and I wrote my original response hastily and didn’t catch my own mistake until later. There is a fair bit of protein in the matrix that makes up the exoskeleton in addition to the chitin; that is, the exoskeleton is a lot more than just chitin. It has all sorts of different layers with different properties and composition (e.g., an outer wax layer), but that topic is pretty tangential here.

The next question was whether the imaginal discs had the same DNA. Yes, every cell in the body has the same DNA, discs included. It’s simply that the discs express a different set of genes than are expressed by the other cells - they’re not like stem cells (i.e., they are not undifferentiated), it’s just that their “program” is a different one. Some of the most fascinating research ever done in cellular/tissue biology has dealt with insect imaginal discs, especially the way they repair themselves if damaged.

The final question (also tangential) about parasites is simple enough: lots of insects make their livings by feeding off of something else - if this feeding is non-lethal, it’s called a parasite, and if it’s lethal, it’s called a parasitoid. Nearly every species of non-marine arthropod has at LEAST one parasitoid that attacks it. That holds true for essentially every insect species that has a caterpillar stage, as caterpillars are generally fairly vulnerable. Your average butterfly or moth caterpillar is potential host for at least 5 or 6 different parasitoids, most of them various types of wasp, but also various flies. Moreover, those parasitoids are THEMSELVES targets of other parasitoids, known as “hyperparasitoids”. So, if you were to spend a few years raising tomato hornworms in your garden, for example, and raised a few hundred of them and kept everything that came out, you’d eventually compile a collection of something like a dozen or more different species of insects. If you also looked at the eggs and pupae, then you’d find still more parasitoids that only attacked those life stages (parasitoids that attack adult insects are fairly rare, though some do exist, like conopid and pyrgotid flies). There are lots of professional entomologists who study this sort of thing, and it’s a truly fundamental part of insect ecology. Some of the life histories involved are truly mind-boggling. Look up the wasp family “Trigonalidae” some time for one such example.


OK, in the article you mention that the organs in the body “basically disintegrate”. What’s the basis for this disintegration? Is it chemical or biological, or some combination? And what would prevent the nervous system from being broken down with the rest of the body?

Articles about this can be read here and here, and here’s the actual study. To dangerously simplify the findings, the tobacco hornworm caterpillars were conditioned to respond to an odor right before they began metamorphosis, and the moths that emerged from the cocoons responded were still averse to the odor.

Believe me, it is a hot topic among the bug nuts. I doubt either presidential candidate will give us a direct response about it.

These bastards just can’t answer a straight question, can they?

After doing a little Internet research I pretty much answered my own question regarding the evolutionary advantage of metamorphosis. Essentially, the advantage is that the various stages don’t compete with each other – it lessens interspecies competition for survival.

But I found one site that said:

a) “More than 80 percent of known insect species undergo complete metamorphosis, suggesting it gives a strong selective advantage across a wide range of habitats. Even more persuasively, the types of insect that undergo complete metamorphosis have many more species than those that don’t, or that only undergo partial metamorphosis.”

b) “The metamorphosis tactic is used by many other types of animals, including many shellfish and amphibians.”

80 percent of insects undergo complete metamorphosis?
Shellfish and amphibians undergo metamorphosis?

Is the definition of metamorphosis more loose than I thought? I know frogs grow from tadpoles to frogs, but I never really thought of that as metamorphosis.

The foregut and hindgut of an insect (the bulk of the digestive system) are part of the exoskeleton, as are all of the tracheae and air sacs that constitute the respiratory system, even though they’re internal - so all of those organs normally are rebuilt from scratch using tissue from the imaginal discs; these organs are broken down the same way the epithelium is broken down - calling it “disintegrating” is a reasonable approximation. The muscles of the adult are built from scratch, as I recall - since they attach to the exoskeleton, that makes sense, and the muscles connected to the pupal exoskeleton are pretty minimal. I’ll admit, though, that I’m not positive about this; it is possible that some, but not all, of the muscles are reorganized rather than being broken down. The gonads are also constructed out of imaginal disc tissue, so about the only other stuff left in the body of the caterpillar are the midgut (a membranous sac), the nerves, and the one dorsal blood vessel. I think the nerves are the only one of these that undergo any significant reorganization (the other two are so simple that no rebuilding is necessary).

I’m curious about the evolutionary history of metamorphosis. In how many lineages did this evolve? Only one, or have separate groups developed it independently? Or is it so complicated that these questions aren’t meaningful?

Or, relatedly, Algernon answers why metamorphosis is advantageous, but how did they get to this complex process? What were the antecedents?