top of page
  • Andy Davis

Effects of captive-rearing on caterpillar anti-predator behavior - an inside look at some preliminar

Hello everyone,

Today I have something very interesting to share with you. It's not a new published study on monarchs, but actually a not-yet-published one that I have recently worked on, and have not even written or submitted to a scientific journal! In fact, today I'm going to show you the actual data that I just collected, before all of this other stuff happens. I'm going to give you the inside skinny, so to speak, on the data before it hits the press. And as the blog title indicates, the results of this project have some implications for the practice of captive-rearing.

Let me start with the big picture here, before getting into the nitty-gritty stuff. I'm going to assume that if you're reading this, then you follow the happenings around monarch butterflies. That means you also are well-aware of the growing fad of captive-rearing monarchs - where people collect eggs and larvae from their backyard milkweed, and bring them inside to rear in a plastic tub. I bet a lot of folks reading this even have monarch larvae somewhere in their house right now! And, if you know all of this, then you also SHOULD know, this practice is generally not condoned by the scientific community, because of the risks of disease spread, genetic concerns, etc. And, now it looks like this practice even affects the navigational ability of monarchs (i.e. the infamous PNAS paper, blogged about here). But...sigh, people are still doing this despite the evidence...

One of the things that most people missed about that PNAS paper was the fact that it highlighted how little we know about the effects of captive rearing on monarch biology, and how disturbingly easily monarchs can be screwed up. At least, that's what I took from this study. In fact, after reading that paper, I got to thinking, I wonder what else can be altered by captive rearing? This one paper seemed to barely scratch the surface. If you think about it, there are so many other things about monarchs that could also be affected too.

So...I got to thinking that one thing that would be interesting to examine is the behavior of monarch larvae. If you consider what the experiences of a "captive-reared" larva are versus one that grows in the wild, you can start to see how this could be an interesting test. When you rear monarchs inside in a plastic container, this inherently involves frequent disturbances, in the form of daily feedings, cleaning the containers, or whatever. Essentially, these larvae are experiencing daily contact with humans, and we don't really know what they think about us. For all we know, they may think that we humans are like giant predators, trying to eat them. A larva living happily in the wild never experiences this. Sure, they may encounter other things in their lives, like other insects on the plants, or bumping up against their brothers and sisters, etc. But we don't really know if the daily "management" of captive-reared larvae is the same degree of disturbance they would experience in the wild. I'm betting it's not, and in fact this small experiment I performed should prove this.

Therefore, I wanted to conduct an experiment designed to evaluate whether monarchs reared in captivity show altered "disturbance reactions" as those that grow up in the wild. In other words, I wanted to see what their behavior would be when they are disturbed by humans. Remember, monarchs reared inside are disturbed by humans on a daily basis. Does this make them more averse to disturbance, or less - i.e. do they "get used to it?"

To do this experiment I capitalized on a fortuitous circumstance. That's a fancy way to say I got lucky. In the lab at UGA, there is always a lot of monarch rearing going on, for various projects. Sometimes, too many larvae are reared for one reason or another, and then we are left with "extras". Last week, we "released" some of these extras onto our milkweed in our yard, so that they could live out their lives in peace in the real world. Ironically, this is kind of the opposite of what people in the captive-rearing world do! At the time, these larvae were pretty young, most were at the early 3rd instar stage.

At the same time, some of the brothers and sisters of those larvae had been reared in the lab for use later. As you might expect, the rearing procedures in the lab generally mirror those in your home. Larvae are kept in plastic containers (5 per container), and fed daily helpings of cut milkweed. The daily feedings also involve cleaning the bottom of the container of frass and old stems. All of this involves gently moving the larvae around a bit. And the important thing is, this happens every day. Meanwhile, the outdoor caterpillars were not experiencing any of this.

The thing that made this very fortunate for me is that these larvae, in both the indoor and outdoor treatment were all from the same family - that means they were all brothers and sisters.

A week later, I went back to the same milkweeds where they were first placed, and I conducted a very simple test of their behavior. It went as follows. I picked up each larva one by one. I held it between my thumb and forefinger for 10 seconds, as the image below shows.

I then placed it on a table for another 10 seconds. All of this was video-recorded. From watching the videos later, I recorded the behavior of each larva after it was picked up. I specifically looked to see if it displayed its characteristic anti-predator defense, which is to curl up in a ball. I'm sure anyone who has handled enough larvae has seen this at one point or another. They curl themselves up tight and won't uncurl until they feel like it's safe to do so. It's something they do when they feel threatened. Here is a picture of one larva doing this on my fingers.

I then went back (the same day) and performed the same test on the indoor-reared monarchs. Since this was done on the same day, both sets of larvae were exactly the same age; all of them were in the fifth-instar stage. And also, I made sure to do this at a time point during the day when the outside temperature matched that of the rearing room inside (82F). This was to ensure that one group of larvae wouldn't be warmer than another, which may lead to behavioral differences.

From each video, I recorded whether or not the larva exhibited this curling behavior, and also which group (indoor or outdoor) they belonged. That was it! In total, there were 19 larvae in the outdoor group, and 27 in the indoor group.

So to sum up thus far, remember that for this experiment, I had behavior data on two groups of monarch larvae. The larvae from both groups were the same age, and from the same family. They had all been measured by one person using the exact same procedure, on the exact same day, and the ambient temperature of both groups was the same. The only difference between the groups is that one spent the majority of its life outside in the real world (about 7 days), and one was reared in captivity, where they were exposed to humans on a daily basis.

Here is what I found:

This graph shows the percentage of larvae that displayed the curling (defensive) behavior during or after being picked up by a human (me). It looks like captive-reared larvae show a much more pronounced defensive behavior than the natural ones did. In fact, they were three times more likely to exhibit defensive behavior than normal monarch larvae were. For those interested, this difference you see in the graph is indeed statistically significant (using a test called a "Chi-square test").

So the results of this mini-experiment were pretty clear. Keep in mind that I am aware this dataset is small. I hope to add more data in the coming months, actually. Then, ideally, these results will be written and submitted to a scientific journal. But for now, from a statistical standpoint, the fact that such a small sample size yielded significant results is really encouraging, because it means there is a real biological pattern here. In other words, this result is probably real.

Now, let's talk about what this means. From this experiment we can conclude that captive-rearing can alter the larval reaction to human contact, although it's not exactly in the direction you might expect. On the one hand, you might think that larvae that are exposed to humans on a daily basis would eventually lose their sense of threat when touched. But, it looks like they actually become more sensitive (or averse) to handling. That's freaky.

I should mention here that I have no idea why this result turned out the way it did. After all, this was a preliminary experiment, tackling a question that has never been addressed before. Of course I don't know why!

I guess I can speculate a bit here, or at least let you know my thoughts... it's possible that the daily bouts of disturbance experienced by the captive larvae triggered repeated physiological stress reactions in them (changes in heart rate, respiration, blood pumping, etc.). In other animals, even in humans, we know that repeated bouts of stress can lead to permanent changes in the way they react to future stressors. This is especially true if the stress experiences happen during the early life of the animals.

There are other possible explanations I could describe, but it's really too soon to be trying to explain this anyway, especially without further experimentation. What we CAN say for now is that this experiment did demonstrate that captive-rearing alters the disturbance reaction of monarch larvae, as I expected. In other words, it changes their behavior. Wow. Who knew?

That's all for now folks. Stay tuned for updates on this project!


Direct link to this blog entry:



The science of monarch butterflies

A blog about monarchs, written by a monarch scientist, for people who love monarchs

bottom of page