How the Venus Flytrap Evolved

The trick to understanding evolution is less about finding good answer to questions, but rather, finding good questions to answer.

Read that sentence twice, because it is very important.

Years ago, Niko Tinbergen developed a method of formulating questions about biology. I'm pretty sure the Tinbergenian method has not been integrated into most science standards and teaching curriculum. It should be.

There are four types of questions one could formulate about a biological system, trait, or observation.

1) Mechanistic. How does this thing work? What cellular processes are involved in a metabolic process, or how do the lever forces work in a joint, or how does a heterotroph get some food into its gut.

2) Ontogenetic. Given the various parts of an organism, how did they arise initially during development? Thinking only of multi-celled animals for a moment, all animals can be divided into large categories that have a common body plan, and that body plan is easily seen in the embryonic state. Looking at a fully formed adult, at any particular organ or part of an organ, we can ask, how did this thing form during the process of differentiating this particular taxonomic category? For example, mammalian pituitary glands are a combination of a bit of what would otherwise be brain, and a bit of what otherwise would be the roof of the mouth. This helps make sense of the pituitary gland.

3) Phylogenetic. How is a particular feature of an organism potentiated or constrained, in its development or function, by ancestry? Birds have either two or four wings (the four winged birds are all extinct). Birds with two wings have two legs. Why not have two legs, two arms, and two wings? Because birds evolve within a taxonomic group that have four limbs, so wings are modified versions of those limbs.

4) Functional. Sometimes called "ultimate" this category of question is largely (but not entirely) about natural selection. What is it about this trait, or this particular form of this train in this species (or sex of a given species) that enhances fitness. This is about the aspect of the trait that presumably caused the trait to spread and become typical, or that caused a particular population (the population with this specific trait) to become more representative over time, due to selection.

A simpler version of this divides all features into two categories, "proximate" and "ultimate." The proximate stuff is the immediate description, what it looks like, how it works, etc. The "ultimate" bit is the functional question (number 4 above). I prefer to keep the four categories in mind.

By the way, there is a thing I call Greg's Rule of Tinbergenian Inquisition (GROTI). This is not a hard and fast rule of nature or logic, but just a common occurrence. If you consider all four Tinbergenian questions in relation to a given trait, one of the four questions will produce a boring answer or a tautology. There are probably deep underlying philosophically interesting reasons for this, but the rule is more important as a guide to teachers. If you want to teach the Tinbergenian method of asking questions about evolution, by example, you will probably need to come up with two or more examples in order to not have one of the four approaches look silly. But I digress.

You want to know how the Venus Flytrap evolved, and some recent research sheds some light on this by looking at the ontogenetic and phyogenetic aspects of known traits.

Venus flytraps are plants that capture and then digest insects (or other small critters). The ultimate reason they do this may have to do with nutrients. Capturing and absorbing the tissues of insects provides nitrogen and some other often rare nutrients. So, the adaptation is to nutrient poor environments. One might guess that this is a trait that was selected for in nutrient poor environments, or one might guess that it is a trait that emerged largely by accident and then allows plants that do this to do better in nutrient poor environments. Or, one might not be too concerned by this distinction and guess that both features of the emergence of a trait are likely to be involved in the evolutionary history of a particular species and its adaptations. But, again, I digress.

So, the Venus flytrap has sensors on it that tell the plant that there is prey present. Then it snaps shut (that is the coolest ting about the plant, but we're not actually going to go into that here). Then a liquid engulfs the prey and digestion happens, then a different liquid is exuded and this facilitates the transport of nutrients into the plant.

Now, think about plants, generally. Plants have all sorts of chemicals in them, or on them, that do all sorts of things. Is there anything about plants in general, about what they normally do, that could provide the genetic (and therefore metabolic or processual) basis for any of these things?


Plants have hairs on the that detect the presence of possible herbivores.

Herbivores may use certain chemicals to break down plant tissue, for ingestion and digestion.

Plants have evolved chemicals that break down those chemicals.

The chemicals that counteract the breakdown of plant tissues probably scare off herbivores, or limit their success, but the can also break down some of the molecules that herbivores are made of.

Meanwhile, plants have genes that are expressed in roots that produce chemicals that facilitate the transport of nutrients from the roots into the rest of the plant.

So, from a recent write-up in Science:

To catch an invertebrate that has blundered into its snare, the flytrap relies on an ancient alarm system. It starts ringing when the victim jostles trigger hairs. The hairs in turn generate electrical impulses that somehow stimulate glands in the trap to produce jasmonic acid—the same signal that noncarnivorous plants use to initiate defensive action against herbivores. Patterns of gene expression in the two kinds of plants confirm the similarity...

...In noncarnivorous plants, jasmonic acid triggers the synthesis of self-defense toxins and molecules that inhibit hydrolases, enzymes that herbivores secrete to break down the plant's proteins. As part of their counterattack, plants also produce their own hydrolases, which can destroy chitin and other components of insects or microbes. In the flytrap, ... [t]ens of thousands of tiny glands make and secrete hydrolases. The trapped invertebrate is drenched in the same digestive enzymes that another plant might use in smaller quantities to ward off an enemy....

Then, plant genes code for chemicals that help absorb the nutrients from the insect.

Experiments showed that many of these genes are the same ones expressed in the roots of other plants. “We looked at each other and said, ‘Yes, it's a root,’” Hedrich says. “It made immediate sense,” because the flytrap draws its nutrition not from soil, but from its prey.

We now have a phylogenetic look at a large part of the Venus flytrap's unique insectivores adaptation. As is generally the case, these novel adaptations are reworkings of pre-existing adaptations.

And, ontogenetic questions arise (but are not directly addressed by this research). How did the genes and their products normally found in roots get into the "stomach" thingie of the flytrap? Did the site of differentiation of root structures move to elsewhere in the plant, or is it just the genes being expressed in different cells? This question came to my mind reading this story, but I'm more of an animal guy than a plant guy. Animals have homeobox genes that control the overall differentiation of tissues, and stem cells that determine and limit, at various levels, what the possible cell types are in a give part of the developing animal. Plants are different. This is one of the reasons that plants can propagate vegetatively and few animals do anything similar.

So I asked Dr. Rainer Hedrich, the flytrap guy who produced, with his team, this research, if this was a case of the movement of root tissue into the business end of the flytrap, or, alternatively, the expression of genes in tissue that are not normally expressed there in a plant.

He told me, "The flytrap develops from tips of Dionaea leaves. So, the trap is a leaf on one side. The inner surface of the trap is covered by a turf of glands, and these glands express genes one otherwise finds in roots. So, the trap is a leaf with root function. Most likely, to serve carnivory, Dionea modified a transcription foactor or promoter of root genes and so directed them into the glands."

So, the story remains one of phylogeny, not ontogeny. The ontogeny part of the story is uninteresting, but the funcitonal, phyogenetic, and mechanistic parts of the story are fascinating.


The paper:

Venus flytrap carnivorous lifestyle builds on herbivore defense strategies. Felix Bemm, Dirk Becker, Christina Larisch, Ines Kreuzer, Maria Escalante-Perez, Waltraud X. Schulze, Markus Ankenbrand, Anna-Lena Van de Weyer, Elzbieta Krol, Khaled A. Al-Rasheid, Axel Mithöfer, Andreas P. Weber, Jörg Schultz, and Rainer Hedrich. Genome Res. Published in Advance May 4, 2016, doi:10.1101/gr.202200.115

Graphic at the top of the post from here. Caption: Venus flytrap with its turf of glands and some gland complexes under the microscope – color enhanced transmission electron micrograph (TEM). B) Cross section of a gland complex showing the three characteristic cell types (Picture: Dirk Becker, Sönke Scherzer, Christina Larisch)


More like this

So there is this plant called dodder that parasitizes other plants, but until recently it was not known how it found the other plants. Recent research suggests that it does so by a form of smell. Dodder is in fact a plant, but when it generates seedlings they will actually wave around towards…
Plants and their herbivores have an interesting and complex relationship. It has been true for quite some time (many tens of millions of years) that terrestrial plants do not move around while animal herbivores do (though I've got friends from Texas who claim that there is a Texan tree that will…
Plants and their herbivores have an interesting and complex relationship. It has been true for quite some time (many tens of millions of years) that terrestrial plants do not move around while animal herbivores do (though I've got friends from Texas who claim that there is a Texan tree that will…
We've all seen the images of receding glaciers and stranded polar bears that accompany talks of climate change. But rising carbon dioxide levels also have subtler and less familiar effects, and may prove to be a boon for many animal groups. Plant-eating insects, for example, have much to gain in a…

There is no question that Tinbergen's "four whys" (causation, ontogeny, function, evolution of behavior) are not a part of any science standards I know (not that I'm and expert...reliable, etc.). There is no evidence of them in the NGSS or the framework for it, either.

I'm not sure either that they are to be found in much college science and if you know that they are I'd love to know. When I think about school biology the developmental readiness of students to deal with a Tinbergian Inquisition seems problematic (AP, IB or an independently developed course?).

An issue, of course, is time, the time it would take to get students to this level and to provide them sufficient experiences to be comfortable with it.

I enjoyed your analysis using GROTI and think it is valuable. I wonder how much it is used in seminars based on papers in behavior. And I enjoyed the prod to think about Tinbergen and his stable of students whose field work remains astonishing. How clever and how focused they were on finding good questions to answer.

Thanks for this post and your posts in general.


By Edward Hessler (not verified) on 17 May 2016 #permalink

Edward Hessler not an expert on science standards?????

I sometimes see the ultimate/proximate dichotomy used in college texts. But the use of this approach comes from the Devore-Wilson school. DeVore and Wilson never worked on anything together, but their students ran their classes and were responsible for much of the curriculum, and this approach was heavily used sometimes explicitly and always implicitly.

I'm pretty sure Jay Phelan's college bio textbook (and bio and physiology) uses this, because he and I are Suaboya, and Suaboya always use the Tinbergan approach.

Very interesting. Any chance of a similar post on pitcher plants?


By Brainstorms (not verified) on 17 May 2016 #permalink

Thanks Greg. I also see the now it seems very occasional use of proximate/ultimate dichotomy.

And thanks for telling me about the Devore-Wilson (EO, I assume) school. It makes me wish there had been some horizontal transfer at Harvard but I know that personalities and politics have ways of intervening.

What I liked about your post (well, many things but in particular) is filling out what we mean when we are teaching an evolution that informs. The 4-whys make you think and as you noted at the outset, what it means to find a good question to answer.

You know an authority on teaching biology at the hs level have you ever asked or discussed this with her? Which is a not too indirect way of saying, if you did, what did she say.

I write a column for the MnSTA newsletter and I'm thinking I might want to use your post for a way into the next one. If I do, I'll attribute you but want to ask whether I have your permission.

Thanks, Greg.

By Edward Hessler (not verified) on 18 May 2016 #permalink

Dean, it would be interesting to look into pitcher plants. But not while breathing!

Ed, I assume she is passing my post out in class this week!

Absolutely, use the post in the newsletter, that would be great.