30,000 Facets Give Dragonflies a Different Perspective: The Big Compound Eye in the Sky

tags: , , , , , , , ,

Male Megaloprepus caerulatus, PANAMA, Rio Chico Masambi, near Gamboa,
11 August 2005; photo by Julie Craves, a Natural Science Research Associate
at the University of Michigan-Dearborn.

Color vision in humans depends upon three light-sensitive proteins, called opsins, that are present in our retinas. Each type of opsin absorbs one color of light in the spectrum. In humans, the colors absorbed by these opsins are red, green or blue. The many wavelengths of light reflected into our eyes from the surfaces of objects around us mix together to provide a rich palette of color. Yet, despite our visual color range, there is a creature with even greater scope; the dragonfly.

Dragonflies are classified into the insect order Odonata (suborder: Anisoptera; different wings) which also includes the physically smaller and slower-flying damselflies (suborder: Zygoptera; twin wings). Dragonflies are characterized by their large multifaceted or "compound" eyes, two pairs of strong transparent wings, and an elongated body. Interestingly, even though they possess six legs (as do all insects), dragonflies cannot walk. Odonates are predatory as adults and the aquatic larvae (nymphs) can deliver an impressively painful bite -- as I discovered when I was a child, catching them bare-handed in a local pond.

All dragonfly species have excellent vision. Each compound eye is comprised of several thousand elements known as facets or ommatidia. These ommatidia contain light sensitive opsin proteins, thereby functioning as the visual sensing element in the compound eye. But unlike humans, day-flying dragonfly species have four or five different opsins, allowing them to see colors that are beyond human visual capabilities, such as ultraviolet (UV) light. Together, these thousands of ommatidia produce a mosaic of "pictures" but how this visual mosaic is integrated in the insect brain is still not known.

These different opsins have very specific arrangements within each ommatidia.

"They are segregated in the compound eye so that the upwards facing eye has only blue and UV receptors, and the downwards facing eye has receptors for longer wavelengths, [such as] green and orange," says Robert Olberg, dragonfly vision expert and professor of biology at Union College in Schenectady.

This patterned concentration of opsin types, particularly those sensitive to blue and UV light, gives special advantages to hunting dragonflies. For example, it is thought that the sky appears to be very bright to a dragonfly, thereby providing a clear background against which small moving prey can be easily detected, according to Dennis Paulson, dragonfly expert and director emeritus of the Slater Museum of Natural History at the University of Puget Sound, in Tacoma.

Are there color-blind dragonflies?

"We don't know," replies Paulson. "There are some [species] that tend to fly only at dusk; perhaps some of them have limited color vision."

Dusk-active dragonflies have sacrificed most of their color vision in favor of increased light-collecting capacity by having fewer, larger facets in their eyes. They also lack all color sensitive opsins except green, which provides the broadest range of light sensitivity for any opsin. As a result, these dragonfly species probably also have a corresponding decrease in overall color perception.

Dragonflies (and bees) have the largest compound eyes of any insect; each containing up to 30,000 facets, and the eyes cover most of the insect's head, resembling a motorcycle helmet. In contrast to a human eye, each facet within the compound eye points in a slightly different direction and perceives light emanating from only one particular direction in space, creating a mosaic of partially overlapping images. Does this mean that dragonflies have 30,000 eyes?

"No," replies Olberg. "It's more like a human having 10,000 to 30,000 photoreceptors spread out across the retina -- but better than that because each facet has several spectral types of receptors."

Dragonflies can also detect the plane of polarization of light, which humans cannot do without the aid of sunglasses. The advantages of this capability are unknown for dragonflies, but other insects are known to use polarized light as a sort of "sky compass" by which they navigate.

Another visual advantage of the multifaceted eye is a dragonfly's acute sensitivity to movement, as anyone who has tried to catch one can tell you.

"Dragonflies can see in all directions at the same time. That's one of many advantages of a compound eye; you can wrap it around your head," explains Olberg. "The spherical field of vision means that dragonflies are still watching you after they have flown by. However, the backward-looking part of the eye has rather low resolution. So, if you want to catch a dragonfly, let it go by you and then swing your net like a baseball bat from behind. If you swing at them while they are approaching they'll usually see the net coming and easily avoid it. They are awfully good at what they do." Olberg concludes.

Many thanks to Professor Robert Olberg, who graduated with a PhD from the UW's Zoology Department, Dennis Paulson, director emeritus of the Slater Museum, and Professor David O'Carroll, who studies insect vision at the University of Adelaide in Australia (after a brief time at the University of Washington, where I met him while I was a graduate student) for allowing me to interview them for this story.

More like this

What has always attracted me to developmental biology is the ability to see the unfolding of pattern—simplicity becomes complexity in a process made up of small steps, comprehensible physical and chemical interactions that build a series of states leading to a mostly robust conclusion. It's a bit…
We mammals have been beaten again. Shrimp have more sophisticated eyes than we do, with the ability to see things we can't, and I'm feeling a bit envious. There are a couple of general properties of light that can be captured and measured with a light detector. One is the amplitude of the light…
Scientist Steven Wiederman from the University of Aukland (shown in image below) has recently published his work suggesting that dragonflies have the ability to focus on a target while blocking out other useless visual information. This is known as selective attention and his work is the first to…
I got a letter from a creationist today, claiming that "Darwinism is falsified," based on an article in Nature. It's kind of amazing; this article was just published today, and the metaphorical digital ink on it is barely metaphorically dry, and creationists are already busily mangling it. It's a…

There are a very small number of humans who have four different types of opsins, and who therefore see many many more colors than the rest of us (and are eagerly sought out by the chemical industry, among others). Because the gene which codes for opsins lives on the X chromosome, and because both parents must carry that gene for this capability to be present, all such "tetrachrome" individuals are female.

Alas, none of them has yet been reported with even a single set of wings.

By Pierce R. Butler (not verified) on 08 Jul 2009 #permalink

Prof Olberg was my advisor when I got my BS at Union College. His talks on dragonfly hunting strategy were amazing. No offense to Dr. Rice, moss is cool too.

Pierce - that is very cool. How would one learn that they see more colors than most people?

wow, pierce, i've also never heard this before. where did you learn this and can you provide any names where i might follow up on this?

Now do an interview with somebody who is knowledgeable about the best eyes in the animal kingdom, those of mantis shrimps. If I recall correctly they may have as many as 16 rhodopsins, and at least as many facets to their compound eyes as do dragonflies. Moreover, they can make a good claim to being the most "intelligent" of the arthropods, whatever that may mean. However ambiguous intelligence is, a mantis shrimp in an aquarium near a desk will spend most of its time watching the person in the desk.

It's definitely an Odonate, but without actually looking up the taxonomic information, I think you have a damselfly, not a dragonfly, in your picture.

By biosparite (not verified) on 08 Jul 2009 #permalink

wow, pierce, i've also never heard this before. where did you learn this and can you provide any names where i might follow up on this

I'm not Pierce, and I don't have the background to critically evaluate it; but here is a paper discussing possible human tetrachromacy:


More on Wikipedia, of course: http://en.wikipedia.org/wiki/Tetrachromat#Possibility_of_human_tetrachromats

I'm an amateur photographer and I have been trying to label my photos correctly and so have spent sometime on this.

I believe damselflies are a suborder of dragonflies.
The dragonflies you are refering to are known as 'true dragonflies' which cannot fold back their wings.

If of interest btw, here is a link to my photos on flickr.


I'm surprised at the line "Yet, despite our visual color range, there is a creature with even greater scope; the dragonfly." Our three rhodopsins (for most of us, that is) may be unusual among mammals. But most vertebrates actually have 4. And it's well documented that many birds can see into the UV. Raptors are known to use UV light to detect rodent's urine trails, which enables them to quickly assess prey densities in unfamiliar habitats. And Blue Tits (I believe it was this species) have been shown to be sexually dimorphic when viewed in the UV. Overall, our visual color range is far from the most impressive out there.

Of course, none of this is meant to take away from the fabulous vision of Odonata.

And then there are the arthropods that can REALLY see... the mantis shrimps or stomatopods. Their eyes have a comparative number of facets to the odonates, but if memory serves, they have 16 rhodopsins. They see deeper into the UV and IR than we do.

Not only that, they can make a good claim for being the smartest arthropods, and make absolutely fascinating pets. Kept in a small aquarium on an office desk they will certainly watch their watcher with what appears to be unbrindled curiosity. They are easily as smart as the typcial undergrad in an introductory biology class...

Now that the visual color range of humans vs. mammals vs. rest of the animal world is brought up, there is one slightly off topic thing that was bugging me.

Now, I've read some stuff about Pepperberg and parrots here, and some time ago I caught a TV documentary featuring one of these parrots competing with a class of children on some tasks (I assume it was one of Pepperberg's parrots, since there was a mention of the deceased Alex). The parrot was competing very well on most tasks but tanked on color recognition.

Not that this was a highly scientific experiment there, but how do they normally ascertain that a green object is the same color as another green object? Compared to birds we are partially color blind after all. They can see a UV component, so they might see different colors where we see the same, or the same color in indoor lighting might appear a different shade in daylight.

Nothing was mentioned in that documentary at least. Is there a way to easily check UV reflectivity? What process is used to screen the objects presented to the parrots for teaching and testing color recognition?

A person would have reason to suspect she's tetrachromatic if she saw clearly distinct hues where others claim, "those are all the same color".

A more (ahem) colorful account is at http://www.cs.utk.edu/~evers/documents/tetraChromat.txt.

Also see (ta-dah!) wikipedia's "Tetrachromat" (link not cited in hopes of avoiding moderation limbo).

OT but not to be neglected under the present circumstances: Grrl - please ask your current companions about the purity of Finns and Swedes.

By Pierce R. Butler (not verified) on 08 Jul 2009 #permalink

thanks everyone for your comments and thoughts on this piece. i originally wrote a shorter piece for a monthly column that i had while i was a grad student, thus, the people whom i interviewed for this story were connected to me either because they were somehow associated with the zoology dept at UW, or were instructors of mine for the master birder program run by the seattle audubon society (dennis paulson). i will spend some time investigating your comments and rewriting this; maybe the rewritten version will be considered for the 2009 installment of Open Lab?

i started this piece by talking about human vision and opsin proteins because when i originally wrote this piece, i was using the undergrad biology course syllabus as a rough guideline for story ideas. i was trying to use that as a starting point to present enough additional information that would appeal to the students and provide a point of interest to the general university community. but a rewrite of this piece specifically for my blog would be aimed more at a bird watching, bird breeding or a bird-appreciating/nature loving audience, since most of my readers fall into this category. (or so i think).

Nigel - thanks for the fix! (Sorry about that extraneous period, folks...)

Grrl - according to the wikipedia article, tetrachromacy is found in many birds, so you shouldn't have to roam too far from your chosen field for more subjects. I can just see your next interview:

Elektra: You never asked!

By Pierce R. Butler (not verified) on 09 Jul 2009 #permalink

Nice blog. Thank you!

If you're thinking of reworking the article, you might want to have a look at that first paragraph. The 3 opsins do not actually have their peak sensitivities at Red, Green and Blue, as your statement implies. For example, OPN1LW encodes for an opsin which has its peak sensitivity around 564â580 nm, which is greenish-yellow, not red. OPN1MW produces a peak at greenish-blue.

Human tetrachromacy can result from the polymorphism of OPN1LW which allows a woman to inherit copies that have slightly different spectral responses.

Since up to 9 copies of OPN1LW and OPN1MW are found on a single X-chromosome, I wonder if some men might not also have this effect.

It's because OPN1SW is on chromosome 7 that it is autosomal and defects occur equally in males and females.

The narrative I have come across is that the earliest mammals were burrowers and were reduced to being dichromatic. When primates became frugivores, the redundant copies on the X chromosome allowed us to be trichromatic again, just not the same kind of trichromats or tetrachromats our pre-mammal ancestors were.

By JohnnieCanuck (not verified) on 10 Jul 2009 #permalink

The possibility of true human tetrachromats is still pretty iffy, but even excluding "color-blind" folks, there are certainly wide variations in human color sensitivity! My own family runs to exceptionally sensitive color vision (we used to call it "Grandpa's eyes"). Naturally, most of his descendants have at least dabbled in the visual arts, and several are amateur artists.

The curious might be interested in the Munsell Color test. (If your monitor does color adjustments, you may want to zero the sliders on that, first.) (And yes, I got a perfect score. ;-) )

By David Harmon (not verified) on 10 Jul 2009 #permalink

As an amateur photographer I discovered that there are different colored eyes amongst dragonflies, at least there are where I live. Just so I can be correct, am I correct in belieiving these actually are dragonflies? If you visit my blog, I have posted several pictures illustrating what I mean. I was really amazed to see a red-eyed and blue-eyed dragonfly today. I'm glad I found your site. mm