A Neotropical Savanna

Here are three red and yellow plants, from left to right, an orchid (Epidendrum radicans), a milkweed (Asclepias curassavica), and a roadside weed (Lantana camara). They all grow in our back yard, although the orchid is there only because I put it there.

If you compare these plants in pairs, you’ll find other features besides color in common. The orchid and the milkweed package their pollen in little sacs called pollinia. The milkweed and the lantana produce nectar (the orchid does not), are not indigenous to Central America (the orchid is), and can be toxic to herbivores. All three are said to bloom year-round, but only the lantana does so in our backyard. All three are said to have the same pollinators, which brings us to the fame this group of plants has among some students and some botanists.

It has been suggested that they belong to a “floral mimicry complex.” But do they? And what is a floral mimicry complex, anyway?

First, the definitions. If you look up mimicry in wikipedia, you’ll find the definition and examples relate to animals. The idea is that a so-called non-harmful species, such as the Viceroy butterfly,

“mimics” -or looks like – a noxious species, such as the Monarch butterfly.

As a result, the Blue Jay, for instance, avoids eating the Viceroy as surely as it avoids eating the Monarch. The Viceroy gets “free” protection. It doesn’t have to take in a toxin from the milkweed, make it harmless to itself, but keep it harmful to its predator. A physiologically expensive proposition.

Seven types of mimicry has been described by zoologists. Two types – Batesian and Müllerian – are of interest here.

Batesian mimicry is named after Henry Walter Bates and is the type exemplified by the Viceroy and the Monarch. The mimic (Viceroy) looks like the model (Monarch) but does not share the trait (toxicity) that makes the model unappealing to the predator (Blue Jay).

Batesian mimicry could work another way – a mimic could imitate a model that provides benefit whereas the mimic does not offer that benefit. In this way we come to floral mimicry, the inverse of animal Batesian mimicry. Flowers, especially orchids, may mimic other flowers or even animals to fool pollinators. If we look at the three plants of interest today, we see that the orchid Epidendrum does not have nectar whereas the other two plants do have nectar. So one idea is that the pollinators of Asclepias and Lantana, having sampled their nectar, might also pollinate the orchid Epidendrum because it looks similar to the milkweed and the lantana.

In my opinion, this is a stretch, but we’ll come back to the idea later.

Müllerian mimicry. This type is named after Fritz Müller. According to wikipedia, “His great discovery was about the advantage one unpalatable species would gain from resembling another unpalatable species.” Thus, all bees and wasps that have alternating black and yellow stripes are instantly recognized as being dangerous and unpalatable.

For plants, Müllerian mimicry again would be the inverse of the zoological definition. If several plants that produced good nectar also looked alike, then pollinators, having learned about the one, would continue to sample the others, and all similar plants would then benefit by being pollinated. The mimics would be recognized as being good, rather than dangerous.

The appeal of the concept of mimicry in the three plants we’re looking at now is that both Batesian and Müllerian types of mimicry might possibly be demonstrated in them. The idea was proposed in 1966 for a course in tropical studies called The Biology of Epiphytes (Bierzychudek). It was hypothesized that the non-nectar-producing orchid mimics the milkweed and the lantana, thereby attracting pollinators to itself. An example of Batesian mimicry in plants.

By contrast, the lantana mimics the milkweed (or vice-versa) so that, by growing in similar areas, the pollinators will visit both plants. An example of Müllerian mimicry in plants.

At bare minimum, for the floral mimicry concept to apply to these three plants, they must have the same pollinators. Indeed, several kinds of nonstinging bees and several species of butterflies have been observed on all three plants. (Bierzychudek) The most common visitor to all three species seems to be the butterfly, Anartia fatima, of the family Nymphalidae. (Bierzychudek).

Anartia fatima images above from Green Hotels of Costa Rica (left) and Insectarium Virtual (right).

Fun
It’s time, then to summarize these concepts. “Asclepias and Lantana are presumed to be Müllerian mimics of each other while the nectarless Epidendrum is thought to be a Batesian mimic of the first two.” (Bierzychudek)

It’s fun to think about, isn’t it? These two types of mimics might be exemplified by three plants, each from a different family.

…if true
In 1981, Bierzychudek published a study that looked closely at the hypothesis. She predicted that more pollen should be removed from dense stands of the milkweed and lantana than from sparse stands. If Müllerian mimicry were a factor in attracting pollinators to the flowers, then this prediction should be true whether the stands were pure or mixed. She also predicted that the orchid should receive more visits when interspersed with the milkweed or with lantana than when growing alone.

She was able to test these predictions by observing the pollinator visits and also by counting the pollinia they collected. As mentioned, pollinia are sacs of pollen. You can find some excellent illustrations of flower anatomy and pollinia at this site, which includes the following helpful sketch from a milkweed (note that the word “pollinium” is singular for pollinia):

Her conclusion: “Although the three species have overlapping ranges and share pollinators, visitation frequencies in stands of different floral composition do not support this hypothesis. Flowers in high-density stands of the “Mullerian mimics” are not visited more often than flowers in low-density stands, so apparent increase in population density through mimicry does not appear to confer an advantage. The “Batesian mimic” is not visited more often when interspersed with the model than when alone.” (Bierzychudek)

So…probably NOT true.

Personally, as I mentioned earlier, I would be skeptical about including Epidendrum in the complex, anyway. Look at them again. The orchid, Epidendrum, is on the left.

True, the colors are similar, but if you scroll through Roxi Steele’s images from Costa Rica, you’ll see several other flowers with this orange/yellow color combination, most notably Psiguria bignoniacea, P. tabascensis, and Caesalpinia pulcherrima. It’s hard for me to believe that a butterfly, at its scale, would confuse an Epidendrum with either the milkweed, Asclepias, or with Lantana.

By contrast, to human eyes, the milkweed and the lantana look similar at first glance. In fact, when a visiting botanist first showed me the Asclepias in our back yard, I said, “Oh, that’s a lantana!” Then I looked again. Now I would never confuse them, but at first, my own mind imposed the known lantana pattern on the unknown new flower.

Still, though, I can’t imagine that a butterfly would confuse them! The flower is about the size of the butterfly itself. I’ve tried thinking about this on a human scale. I suppose a dining room table with a few plates of food on it would be relatively the same size to me as these flowers are to a butterfly. If I walked into a dining hall where food was on the tables, what would I be fooled by? From a distance, plastic food. That would be Batesian mimicry – plastic food offers me no reward. Hmm. Actually, it’s the tables that should mimic, or not, each other. Yellow tables mean appetizers; red tables, desserts. That would be Müllerian mimicry. Ugh. It’s getting hard to carry through on this analogy. Oh, well. You get the idea.

1981 to 2007

At any rate, what interests me is that 26 years after the publication of Bierzychudek‘s paper, some botanists still refer to the Epidendrum, Asclepias, Lantana floral mimicry complex. It was mentioned, with a caveat, in my deeply used book, A Guide to Tropical Plants of Costa Rica. It was mentioned by my visiting botanist friend. I’ve found it in a PowerPoint presentation that can be downloaded by students taking advanced botany at a prestigious university.

When I wrote to Paulette Bierzychudek about this conundrum, she modestly replied: “… it’s entirely possible that I’m wrong – that some other study has been or will someday be done that demonstrates that mimicry takes place. I don’t know of any. Still, myths die hard.”

A pretty good description of the situation was published in a paper entitled “Floral mimicry: a fascinating yet poorly understood phenomenon.” (Roy & Widmer) The first sentence of their abstract is: “Flowers of different species that resemble each other are not necessarily mimics.” How sensible!

Floral mimicry would be fun, in this case, if it were true.

Update 1: I’ve just learned that Bitty Roy (of Roy and Widmer, next-to-last paragraph) obtained her PhD under the direction of Paulette Bierzychudek. In the interest of full disclosure.

Update 2: In case I’ve given the misleading impression that all floral mimics resemble other plants that give benefit, it is at least possible that some floral mimics resemble other plants with undesirable traits. It has been observed, for instance, that both Asclepias and Lantana are unpalatable and toxic. Since they both occur in pastures, their resemblance may provide protection from grazing. (Bierzychudek). This possibility had not been tested as of 1981 and I do not know of it having been tested as of today.