Why do animal species vary so much in the shape of pupils in their eyes?

Here’s a picture of some animal pupils from National Public Radio. The website showed them because they’re the subject of a new paper by M. S. Banks et al. in Sciences Advances (reference below; free download). The paper asks an interesting question: why do the pupils of vertebrate species vary so widely in shape? (Supplementary material, including database and movies, can be found here.)


(From NPR): Can you guess which eyes belong to what animal? Top row, from left: cuttlefish, lion, goat. Bottom row, from left: domestic cat, horse, gecko. Top row: iStockphoto; bottom row: Flickr

Now the paper is long and complicated, with lots of simulations and math, but the upshot is pretty simple: animals tend to have vertical pupils if they are nocturnal “ambush” predators and aren’t too tall, while animals that have horizontal pupils tend to be prey that need to detect predators approaching them and also to scan the surrounding terrain to find a good escape route.

These conclusions come from surveying 214 terrestrial species and gathering information about their pupal shape, when they were active, and their mode of foraging. This was combined with optical and mathematical analysis to determine the optimal shape to see under different conditions.

The plot below shows three modes of foraging (herbivores, active predators who search around for their prey, and ambush predators who remain stationary before attacking)—all plotted against pupil shape. You can see that there’s a strong relationship between pupil shape and foraging mode: herbivores tend to have horizontal pupils, active predators circular and subcircular pupils, and ambush predators vertical pupils. (Each species is a separate data point.)

Screen Shot 2015-08-10 at 7.55.35 AM

(A) Different pupil shapes. From top to bottom: vertical-slit pupil of the domestic cat, vertically elongated (subcircular) pupil of the lynx, circular pupil of man, and horizontal pupil of the domestic sheep. (B) Pupil shape as a function of foraging mode and diel activity. The axes are pupil shape [vertically elongated, subcircular (but elongated vertically), circular, or horizontally elongated] and foraging mode (herbivorous prey, active predator, or ambush predator). Each dot represents a species. Colors represent diel activity: yellow, red, and blue for diurnal, polyphasic, and nocturnal, respectively. The dots in each bin have been randomly offset to avoid overlap.

As the statistical analysis below show (and the graph above, in which the green dots are absent in vertical ambush predators) there’s also a relationship between pupil shape and activity period. That’s also shown in the following statistical analysis. The P values, which show the strength of the relationship (the lower the P‘s, the stronger the relationship between pupal shape and activity period or “activity”),imply that foraging mode is more important a determinant of shape than is activity period. In fact, there’s no relationship at all between circular shape and activity period.

Screen Shot 2015-08-10 at 7.55.47 AM

(C) Results of statistical tests on the relationship between foraging, activity, and pupil shape. Multinomial logistic regression tests were conducted with foraging mode, activity time, and pupil shape as factors and genus as a covariate. Relative-risk ratios were computed for having a circular, subcircular, or vertical-slit pupil relative to having a horizontal pupil as a function of foraging mode or diel activity. Activity time proceeded from diurnal to polyphasic to nocturnal. Foraging mode proceeded from herbivorous prey to active predator to ambush predator. When the relative-risk ratio is greater than 1, the directional change in the independent variable (foraging or activity) was associated with a greater probability of having the specified pupil shape than a horizontal pupil.

The bulk of the paper is taken up with a complicated analysis of why different foraging modes should select (evolutionarily) for pupils of different shape. I’ll summarize the results very briefly; readers who seek further analysis should consult the original paper.

Vertical slit pupils.  It turns out that having a vertical slit in low light gives ambush predators an advantage in calculating the distance to a prey item. (Remember, these predators must accurately gauge the distance to a prey item before they strike.) The calculations suggest that a vertical slit is better at maximizing the blur of objects that aren’t in focus. That is, it helps the predator gauge the distance to a prey by seeing which aspects of the landscape are blurred and which are not. The vertical pupil also aids in stereopsis: the comparison of images from different eyes that is also used to judge distance to prey. In general, then, distance-judging, which is vital if an ambush predator is to eat, is maximized in low light by having a vertical slit. For such a predator, those that can best judge distance to prey are those that get more noms, and thus leave more offspring. And that, O Best Beloved, is why your small, ambush-predator cat has eyes with vertically slit pupils.

That said, the advantage of having a vertical slit, for complicated optical reasons, diminishes as the eye gets higher off the ground. So the authors made and tested a prediction made from theoretical considerations:

We predict, therefore, that shorter frontal-eyed, ambush predators will be more likely to have a vertical-slit pupil than taller animals in that niche.

We evaluated this prediction by examining the relationship between eye height in these animals and the probability that they have a vertically elongated pupil. There is indeed a striking correlation among frontal-eyed, ambush predators between eye height and the probability of having such a pupil. Among the 65 frontal-eyed, ambush predators in our database, 44 have vertical pupils and 19 have circular. Of those with vertical pupils, 82% have shoulder heights less than 42 cm. Of those with circular pupils, only 17% are shorter than 42 cm.

Nearly all birds have circular pupils. The relationship between height and pupil shape offers a potential explanation. A near and foreshortened ground plane is not a prominent part of birds’ visual environment. The only birds known to have a slit pupil (and it is vertically elongated) are skimmers [Rynchopidae]. . .

Foxes, for example, have vertically slit pupils, but the taller wolves have round ones.

Horizontal slit pupils. Based on a similar analysis, the authors found that, in theory, terrestrial animals that are prey should have horizontally slit eyes. For that shape of pupil enables them not only to get a wide look at the horizon around them, thus detecting the presence of predators, but also scope out escape routes by surveying at the landscape. As they say:

Most terrestrial lateral-eyed animals are prey, so their adaptive strategy is to detect predators approaching along the ground and to flee quickly to avoid capture. The visual requirements for this strategy are striking. On the one hand, these animals must see panoramically to detect predators that could approach from various directions. On the other hand, they must see sufficiently clearly in the forward direction to guide rapid locomotion over potentially rough terrain. In both cases, the regions of greatest importance are centered on or near the ground.

Their analysis shows that horizontally elongated pupil gives a better panoramic view of the ground.  This led to another prediction. If this is the reason for such a pupil, then if the animal changes its head alignment, as when grazing, it should still keep its eyes horizontal so the pupils align with the horizon. And this is what they observed when looking at grazers, which show this “cyclovergence.”

Conclusion: The authors have made a good hypothesis that seems to explain well the diversity of pupil shapes among animals based on both their foraging mode and time of activity.

I see one potential problem: the correlations are based on many species that are closely related (there are many snakes, for example), so each data point does not necessarily represent an independent act of natural selection molding pupal shape, which is what their correlation in effect assumes. In reality, animals could have inherited their pupil shape from a common ancestor and not evolved it independently, so using, say, ten related snakes is not the same as documenting ten independent events of evolution. This problem, one of “phylogenetic inertia,” reduces the authors’ statistical power. (As an example, we can’t claim that all cat species independently evolved sharp canine teeth to kill their prey and rip apart its flesh. So you can’t use every cat species as an independent test of whether sharp canines are associated with ambush predation.)

They recognize this problem and try to overcome it by giving instances in which animals have diverged from the pupil shape of their ancestors in the predicted direction based on changes in lifestyle. The ancestral canid, for instance, is thought to have been an ambush predator active at several times of day, and having subcircular pupils. The authors note that vertical slit and truly circular pupils evolved independently twice each within the canids, and in the direction expected from their activity periods and hunting patterns.

This does show that “phylogenetic inertia” can be overcome, and that the data points are to some extent independent. But we don’t know how independent, for showing some independent evolution doesn’t tell us how often phylogenetic inertia has rendered other points non-independent.  As far as I can see, the authors have recognized the problem of non-independence and dealt with it the best way they can, but haven’t fully overcome the problem.

What they need is an analysis that uses only those data points known to represent independent cases of evolution (Allen Orr and I did this in our work on speciation in Drosophila). I’m not sure it’s possible in this dataset, but until it’s done, the statistical strength of their data remains in question. I think the authors are probably right in their conclusions, and I really like the paper, but I’d also like to see a statistical analysis that is based on “phylogenetically corrected” data.


Banks, M.S. et al. 2015. Why do animal eyes have pupils of different shapes? Science Advances, Aug. 7, 2015, online.


  1. Chris Walker
    Posted August 10, 2015 at 11:43 am | Permalink

    Very cool topic. Thanks, as always, for the breakdown. I also really like that you include constructive criticism of the papers that you highlight here. You bring up points that many of us laypeople wouldn’t necessarily consider.

    • Benjamin Branham
      Posted August 10, 2015 at 12:17 pm | Permalink

      Ditto. My favorite type of post, by far! Thank you x2!

      • Posted August 10, 2015 at 12:22 pm | Permalink

        Sadly, not too many people read these (or maybe they do, but they don’t comment).

        • tomh
          Posted August 10, 2015 at 12:42 pm | Permalink

          Don’t be discouraged. I’m sure a lot of people are like me, interested in and read the post, but have nothing constructive to add, so don’t comment.

          • DrBrydon
            Posted August 10, 2015 at 12:43 pm | Permalink

            Indeed. I don’t necessarily want to show my ignorance by saying “neato.” (Neato post by the way!)

            • MarkT
              Posted August 10, 2015 at 1:09 pm | Permalink

              After reading the review here, I feel I learned more than I would elsewhere. To test this, I read coverage of this at WaPo and SlashGear and neither article had as much depth as this one.

              The other articles did not have any constructive criticism either. I like that PPC’s coverage contained both the key points of the study and insight on how science works 🙂

              • frednotfaith2
                Posted August 10, 2015 at 8:54 pm | Permalink

                I’ve gotta admit in all the decades I’ve been around domestic cats and seen big cats, either in photos, on tv or at zoos, I never noticed that unlike small cats, the big cats have round pupils, more in common with our eyes than their feline cousins. And it makes perfect sense that there must be a particular reason for that as well as for the differences in the eyes of other vertebrates. And all explainable by evolutionary science rather than “well, that’s just how god made them” nonsense.

          • EvolvedDutchie
            Posted August 10, 2015 at 1:18 pm | Permalink

            This applies to me too.

          • eric
            Posted August 10, 2015 at 2:12 pm | Permalink

            Yes that’s me on a lot of Jerry’s posts. Everyone discusses politics and religion (and toilet paper hanging), but once you start talking about the best leather for cowboy boots, photo lens selection in nature photography, or the evolution of goats, I’m probably just going to lurk.

            • slandermonkey
              Posted August 11, 2015 at 10:38 am | Permalink

              I do enjoy the science articles. When you read about this sort of thing in the popular press it is generally simplified and sensationalized. And you don’t get an insightful critique from an expert.

              • HaggisForBrains
                Posted August 11, 2015 at 11:34 am | Permalink


        • Benjamin Branham
          Posted August 10, 2015 at 1:40 pm | Permalink

          That is indeed the case for me. And I see a string of similar cases below. Sadly, the quality of the posts seems to almost be a deterrent from making a non-contributing comment. Like commenting directly in a science journal. I wish you had a “number of views” counter to provide a direct measure of each posts visibility. I also share the info with my daughter, and she loves it!

          • Mark Sturtevant
            Posted August 10, 2015 at 3:38 pm | Permalink

            I have also wondered about the # of views. I expect it is commendably high.

        • Adam M.
          Posted August 10, 2015 at 2:26 pm | Permalink

          I’ll just chime in and say that I’m more likely to read this kind of post than any other, but if your numbers indicate fewer people reading, this might be part of the reason: I’m much less likely to comment on these types of posts, so I’m much less likely to actually visit your website. I can read the text of the article from my RSS reader, but that doesn’t count as a page view in your stats.

          It is a very interesting article!

        • Posted August 10, 2015 at 2:39 pm | Permalink

          It’s posts like these that keep me here. I saw that plot of shape v mode and thought, “oh, wow….”

          On the subject of phylogenetic inertia…what about arthropods? Might there be some way to extrapolate the principles to compound eyes?


        • aldoleopold
          Posted August 10, 2015 at 6:11 pm | Permalink

          I’ll try to comment on more of them – your science posts keep me coming back!

        • Dominic
          Posted August 11, 2015 at 4:04 am | Permalink

          It is hard to make a comment outside one’s area of knowledge or expertese, unless it be to pose a particular technical question. I have only internet at work these days (i prefer to be able to read without distraction at home) so sometimes have little time to absorb these posts & consider what I might ask if I came back a day later… if you see what I mean!

          • Dominic
            Posted August 11, 2015 at 4:22 am | Permalink

            EXPERTISE! How did I do that?!

        • dale
          Posted August 11, 2015 at 1:07 pm | Permalink

          I would be in the — I read these, but don’t comment category. I like these articles the best simply because I am learning. But I am like the quite student in the background, taking it all in while trying not to be noticed (I don’t want to look the fool)

          Yet, I have no problem looking the fool when it comes to politics. I’m really good at that.

        • Posted August 12, 2015 at 10:31 am | Permalink

          Sadly I read many of these articles but never think to comment. maybe a simple thank you will let the autors know they are read and appreciated.

  2. Posted August 10, 2015 at 12:02 pm | Permalink

    The yellow mongoose also has very strange-looking eyes.

    • Randy Schenck
      Posted August 10, 2015 at 12:12 pm | Permalink

      The mongoose is a daytime animal if I remember correctly. I know they imported them to Hawaii long ago thinking they would take care of the rat problem but they didn’t because the rat is the nighttime.

      The mongoose also has extreme reflexes. One of the not so nice tricks they like to show in the far east is a mongoose in a cage with a python. Everyone thinks the mongoose is a goner but it kills the snake.

      • darrelle
        Posted August 10, 2015 at 1:37 pm | Permalink

        Years ago I watched a documentary about research on venomous snakes in which while caring for the snakes being studied a guinea pig that was fed to one of the snakes killed the snake instead. The researchers were surprised. They then fed the same guinea pig to another of the snakes. Again, the guinea pig instead killed the snake. The researchers then retired the guinea pig to a well deserved life of mascot luxury.

        • frednotfaith2
          Posted August 10, 2015 at 9:02 pm | Permalink

          Score two for the guinea pig! Of course, snakes have to eat too. Now the g.p. can tell its grandkids about how he twice was nearly sacrificed to the evil snake demons but managed to turn the tables on them.

      • Posted August 10, 2015 at 6:26 pm | Permalink

        So why do yellow mongooses have horizontal pupils while meerkats have those cute round eyes? The two species are closely related and even share the same environment.

        • Randy Schenck
          Posted August 10, 2015 at 7:15 pm | Permalink

          Not qualified to answer but the mongoose are closely related to cats if that helps. The meerkats live in underground tunnels and their predators are primarily eagles.

  3. eric
    Posted August 10, 2015 at 12:21 pm | Permalink

    Very interesting bit of science.

    I wonder if this also explains why my cat (and I believe cats in general) are so cautious about new environments; their eyes aren’t adapted to do quick ‘safety’ scans of the surroundings.

    • John Scanlon, FCD
      Posted August 10, 2015 at 12:34 pm | Permalink

      Cats are one of the groups analysed in the paper, because pupil shape varies from round to vertical-slit, and they also vary greatly in body size and amount of nocturnal activity, while (usefully) keeping other aspects of foraging behaviour relatively constant.

  4. Jacques Hausser
    Posted August 10, 2015 at 12:43 pm | Permalink

    The splitted eye is also very useful for polyphasic predators (active both night and day), which have to face very contrasting levels of light. From the same paper:
    “Constriction of circular pupils is achieved by ring-shaped muscles, whereas closure of slit pupils involves two additional muscles that laterally compress the opening, allowing much greater change in area. For example, the vertical-slit pupils of the domestic cat and gecko undergo area changes of 135- and 300-fold, respectively, whereas humans’ circular pupil changes by ~15-fold.”

    • eric
      Posted August 10, 2015 at 2:13 pm | Permalink

      More coolness! Thanks.

      • eric
        Posted August 10, 2015 at 3:47 pm | Permalink

        OTOH, this means that sunglasses on a cat are entirely unnecessary. Too bad for you kittehs, the coolness of your eye musculature means you lose out on the coolness of wearing shades.

  5. EvolvedDutchie
    Posted August 10, 2015 at 12:48 pm | Permalink

    I’m surprised to see those papers publicly available. In my country, those papers are part of an online library that can only be accessed by students, university teachers or paid members. There are efforts to make science papers available to the general public, but – as bureaucracy dictates – progress is slow.

    • Diane G.
      Posted August 11, 2015 at 3:06 am | Permalink

      That’s still a big problem in the States, too.

      • visionGuy
        Posted August 11, 2015 at 7:22 am | Permalink

        Yes, access to the scientific literature in the US is not as free as it ought to be, given that we all paid for its creation, but there is real progress in the right direction.

        • Diane G.
          Posted August 12, 2015 at 3:55 pm | Permalink

          Indeed, thank goodness!

  6. darrelle
    Posted August 10, 2015 at 1:08 pm | Permalink

    A very interesting study, and thank you for the excellent post Jerry. I’m looking forward to digging into this one in a little more detail when I have the time.

    I am also looking forward to reading this post with the kids. They will love it! The general topic is right in their wheel-house, and they will get more exposure to critical thinking and the methods of science, especially from your comments on the paper.

  7. Dee
    Posted August 10, 2015 at 1:58 pm | Permalink

    Like some of the other commenters, I don’t generally comment on these posts because I don’t have the background to add anything of substance. But I enjoy them a lot, especially since I wouldn’t get to see this kind of thing otherwise.

    I also really appreciate your willingness to summarize and comment on these papers. That helps me understand what’s going on and what the limitations might be. It’s a fair amount of effort on your part, but ends up being really cool for me.

    • Rod
      Posted August 10, 2015 at 2:51 pm | Permalink

      My sentiments exactly.
      This was a subject that I asked my Biology-major daughter about and she came up blank….it doesn’t seem to be addressed in the standard biology curriculum.
      Great post!

  8. Posted August 10, 2015 at 2:01 pm | Permalink

    Interesting that the only bird with vertical pupil shapes is the black skimmer, which ambushes it’s prey (fish) by flying low over the water skimming it’s lower beak. This seems to suggest that it is viewing its predatory landscape in the same way that a small mammal would.

    • Posted August 10, 2015 at 6:38 pm | Permalink

      You beat me to it about the skimmers. Skimmers must be an example of divergence from the ancestral form due to their feeding behavior.

  9. RobbertPatrison
    Posted August 10, 2015 at 2:07 pm | Permalink

    Is it proven that vertical slits gives a meaningful optical advantage for depth perception? Blur and Stereo-viewing are both optical technicalities that should be measured before such statement can be made. I would not be surprised if the optical advantage of is insignificant. In low light the pupils open so are not in a slit-configuration. Only in high-light would there be such advantage, if any.
    If the animal needs to hunt both during day and night, then it is important to be able to change the iris opening over a larger range. As the previous commenter noted, the split configuration in cats seems to allow for an order of magnitude larger size range than the circular pupil in humans. That is probably a more plausible explanation.

    • darrelle
      Posted August 10, 2015 at 3:10 pm | Permalink

      But that explanation does not account for horizontal vs vertical slit configurations and their correlation with different animal foraging behaviors.

      It also does not account for the trend in animals of similar foraging behavior to have vertical slit configuration pupils below a certain height, but round pupils above a certain height.

      Of course it is a virtual certainty that the order of magnitude larger size range that you cite, and that Jacques Hausser mentioned in comment 4 above, is part of the answer, but it seems as certain that it is not the complete answer.

      • RobbertPatrison
        Posted August 10, 2015 at 8:15 pm | Permalink

        The premise needs to be proven than an iris with a slit has better optical properties than a round iris for depth perception and/or for wider angle of view. If there are none (which I postulate), then any correlation between foraging behavior or size is more likely to be an evolutionary coincidence. It could also be bias in the selection of species that are used in the correlation.

        • rickflick
          Posted August 10, 2015 at 8:31 pm | Permalink

          I imagine this could be modeled either physically or mathematically. The proof of the pudding is in the tasting.

        • darrelle
          Posted August 11, 2015 at 9:00 am | Permalink

          I haven’t had opportunity yet to look at the actual study in detail, but from the OP describing the paper . . .

          “This was combined with optical and mathematical analysis to determine the optimal shape to see under different conditions.

          . . . and . . .

          “That said, the advantage of having a vertical slit, for complicated optical reasons, diminishes as the eye gets higher off the ground.

          Those statements by the OP suggest that optical modeling was a primary aspect of this study. They then made some predictions based on their optical analyses and collected some observational evidence that seems to support those predictions.

          Definitely not definitive, but not bad. The OPs conclusion rightly points out that their data set is not as large a sampling as the numbers suggest because of a relatively high level of relatedness, leaving more room for bias.

          I am not suggesting your hypothesis is implausible, just pointing out that the study seems to have included optical analysis and modeling in just the way you say needs to be done.

  10. Posted August 10, 2015 at 2:12 pm | Permalink

    I’ve always found this topic fascinating, though admittedly in a more aesthetic sense. As humans we’re so centered on our own perception that the concept of having a field of vision such as that of the horse, which has a FOV of nearly 360d, is almost alien. What we humans see is that which is “in front” of us. That which is behind is us unseen. That’s why some people do things *right under your nose* or *in front of your face* while other, naughty people, do things *behind your back.* There’s a dichotomy of experience there that, were we outfitted like our horsey friends, wouldn’t even be a concept, or not so strong of one.

    I’m rambling, but thank you for posting this. Super interesting 🙂

  11. Posted August 10, 2015 at 2:19 pm | Permalink

    This is very cool, thank you!

    I really have always wondered about this.

    • rickflick
      Posted August 10, 2015 at 2:33 pm | Permalink

      Me too.

    • Wunold
      Posted August 10, 2015 at 4:55 pm | Permalink

      Me too!

      I normally don’t like and thus don’t make “Me too!” comments. I did this one to lift the veil from this one more silent admirer of these kind of posts. 🙂

  12. Rob
    Posted August 10, 2015 at 2:45 pm | Permalink


    The Prairies Rattlesnakes I encounter have vertical slits, while the Gopher Snakes have round pupils.

  13. Mark Sturtevant
    Posted August 10, 2015 at 3:36 pm | Permalink

    The vertical pupils of nocturnal predators is well known, but I did not know about the use of other pupil shapes. This is cool.

  14. Tulse
    Posted August 10, 2015 at 3:40 pm | Permalink

    This is a very minor statistical quibble, but I think that description of p-values (“the lower the P‘s, the stronger the relationship between pupal shape and activity period or “activity”) isn’t precisely accurate. The p-value tells us how confident we should be that the stated degree of relationship actually exists — it doesn’t tell us about the strength of the relationship itself. (The statistical relationship between pupil shape and activity period could be very weak, and we could be very confident of that, or alternatively, the statistics could say that the relationship is strong, but that our confidence in that claim should be low.) If I am reading the data correctly, it is the relative-risk ratio that actually tells us the strength of the association between pupil type and activity.

    • Posted August 10, 2015 at 7:55 pm | Permalink

      Agreed! Thanks.

    • TJR
      Posted August 11, 2015 at 4:44 am | Permalink

      You beat me to it.

      I disagree that its a minor point, though.

    • Steve Walker
      Posted August 11, 2015 at 5:50 am | Permalink

      A more pithy way to express this issue is:

      ‘a difference in statistical significance is not itself necessarily statistically significant.’


      • Posted August 11, 2015 at 1:00 pm | Permalink

        I struggle mightily with questions of statistical significance. If I understand the point Wegleman is making, it’s the same as Munroe does here:


        …and, if it’s not the same, then I don’t understand the point…which would hardly be surprising….


        • Tulse
          Posted August 11, 2015 at 1:34 pm | Permalink

          I believe the point being made is indeed similar, although I think it is easier to understand in Munroe’s case. He cleverly presents 20 experiments, one of which shows a significant result at P < .05. In other words, with that p-value, a difference of the size the experiment found would be expected by chance 5% of the time, or 1 in 20. That's why cherry-picking statistical results is problematic, and why it is important to correct for multiple comparisons in these kind of situations.

          • Posted August 11, 2015 at 2:27 pm | Permalink

            Thanks. Makes me feel marginally less unconfident….


    • visionGuy
      Posted August 11, 2015 at 7:27 am | Permalink

      Great point, though I do not consider this a “minor statistical quibble.” In fact, Jerry’s shorthand way of communicating what p-values are is an error, plain and simple. Unfortunately, that same error is committed by way too many researchers.

      • Tulse
        Posted August 11, 2015 at 8:22 am | Permalink

        Well, yes, I didn’t want to seem rude, but this mistake is really just wrong, although as you note it is common to see, from undergrads all the way up to working researchers.

  15. Posted August 10, 2015 at 5:12 pm | Permalink

    Interesting paper. I’m surprised that they didn’t include some monitors (Varanidae) in the study.

    • rickflick
      Posted August 10, 2015 at 5:31 pm | Permalink

      Web images of the monitor seem to show a round pupil. Does this support the hypothesis?

  16. Dan Fromm
    Posted August 10, 2015 at 5:33 pm | Permalink

    Interesting. I wonder what their model would say about Loricariid catfishes omega shaped pupil.

    • winewithcats
      Posted August 10, 2015 at 7:28 pm | Permalink

      Or the peacock mantis shrimp.

      • Diane G.
        Posted August 11, 2015 at 3:12 am | Permalink

        Well it was limited to vertebrates. 🙂

  17. Kevin
    Posted August 10, 2015 at 5:41 pm | Permalink

    Great post appears the authors need to do a little bit more homework to make this a robust study and otherwise really cool.

  18. keith cook + or -
    Posted August 10, 2015 at 5:45 pm | Permalink

    Great post of the ‘interesting and I’ve always wondered why’ the variety of eyes. Before this, I have just thought it weird and interesting.

  19. Posted August 10, 2015 at 6:05 pm | Permalink

    Very nice. This has been a question I had for some time comparing the eyes of my family’s pets. Thank you for presenting and be assured that even if these posts don’t have many comments they get plenty of eyeballs.

  20. aldoleopold
    Posted August 10, 2015 at 6:07 pm | Permalink

    Great analysis, Jerry. I came away with the same thought – it would indeed be nice to see statistical analysis performed with “phylogenetically corrected” data.

    Function of pupil shape has long been debated. In 1919 Hartridge hypothesized that function of a vertical slit pupil (i.e. domestic cats) revolved around their tree-climbing/dwelling habits. Contours of trees and branches are mostly vertical, so if illumination of the image formed on the retina could be increased by decreasing the definition of the horizontal planes, it would be advantageous. Doesn’t quite hold up though considering other predators like alligators hunt along horizontal planes, yet have vertical slit pupils.

    Brings a bit of Yeats’ poetry to mind…
    “Does Minnaloushe know that his pupils
    Will pass from change to change,
    And that from round to crescent,
    From crescent to round they range?
    Minnaloushe creeps through the grass
    Alone, important and wise,
    And lifts to the changing moon
    His changing eyes.”

  21. ThyroidPlanet
    Posted August 10, 2015 at 8:19 pm | Permalink

    “And that, O Best Beloved, is why ….”


    • Posted August 10, 2015 at 9:15 pm | Permalink

      Just so you know, I got that referenced as well.


    • Diane G.
      Posted August 11, 2015 at 3:15 am | Permalink

      I smiled when I got to that. 🙂

      • rickflick
        Posted August 11, 2015 at 5:18 am | Permalink

        I had to Google. 😎

        • Diane G.
          Posted August 12, 2015 at 3:32 pm | Permalink

          Oh, you must read the Just So Stories!

          “Go to the banks of the great grey-green, greasy Limpopo River, all set about with fever-trees…”

  22. mordacious1
    Posted August 11, 2015 at 1:48 am | Permalink

    Snakes’ eyes have always interested me. Venomous snakes tend to have pupils like a cat and non-venomous snakes tend to have rounded pupils. Both types of snakes are (obviously) low to the ground and both are predators. Why the difference? Is it because venomous snakes tend to strike more above their body height than non-venomous snakes?

    My friends who hunt rattlesnakes tell me that snake boots generally are from 16″ to 19″ in height, since the snakes usually strike at mid calf and unless the snake is on a ledge or log, he won’t go much higher. My corn can strike that high, so maybe strike height is not a factor.

    • John Scanlon, FCD
      Posted August 11, 2015 at 7:37 am | Permalink

      There’s been quite a lot of evolutionary change in pupil shape within the elapid snakes, with multiple origins (and several reversals) of the vertical ellipse from the round state, and at least a couple of species are polymorphic round/oval. Approximately all the cobras, seasnakes and large, active Australian elapids have round pupils, with vertical ones appearing (and disappearing)in various smaller nocturnal, sheltering or ambushing groups. It might be quite weakly selected, or picking up on different aspects of behaviour that we wouldn’t yet know much about.

      • Tulse
        Posted August 11, 2015 at 8:20 am | Permalink

        What’s especially interesting is that the variation within related species suggests that pupil shape is a relatively easy thing to change genetically.

      • Diane G.
        Posted August 12, 2015 at 4:24 pm | Permalink

        Taken recently in my yard of a small, nocturnal one:

        IMG_8212 Massasauga


        • Diane G.
          Posted August 12, 2015 at 4:25 pm | Permalink

          Ha! You can just see the eye if you click on it. She wasn’t posing very well.

  23. Mike
    Posted August 11, 2015 at 2:38 am | Permalink

    Vertical/horizontal pupils also allow the retention of good chromatic vision when contracted. Allowing each part of the lens (which tend to have different colour refractive indices from the edge of the lens to the centre) to focus the light. If contracted circularly the image would have significant colour aberration.

    There’s also an hypothesis that the weird pupil shapes in cephlapods allow colour vision even though they only have a single photoreceptor type by taking advantage of the degree of chromatic aberration. This would account for the paradox of apparently monochrome vision with a wildly colourful mating displays etc.

    • Diane G.
      Posted August 11, 2015 at 3:28 am | Permalink

      That is really interesting, thank you!

    • rickflick
      Posted August 11, 2015 at 5:20 am | Permalink

      Good stuff!

  24. Nick
    Posted August 11, 2015 at 2:51 am | Permalink

    Thanks for posting this, Jerry. The science topics on this website are always my favourites. (I’m a non-commenter, mostly, so I just wanted to add my voice to the list of those above)

  25. Dominic
    Posted August 11, 2015 at 4:01 am | Permalink

    Very nice! Clever bit of work.

  26. Torbjörn Larsson, OM
    Posted August 11, 2015 at 4:04 am | Permalink

    A real eye opener, despite all the remaining concern (phylogenetic independence; real mechanism).

    Thanks, Jerry!

  27. Richard Bond
    Posted August 11, 2015 at 4:05 am | Permalink

    …horizontally slit eyes. For that shape of pupil enables them not only to get a wide look at the horizon around them…

    I am not sure that this is correct. If I remember correctly, and to a first approximation, the field of view is independent of the aperture. After all, the human FOV does not increase as our pupils expand from 2 mm in bright light to 7-8 mm in poor light. What does vary (inversely) with aperture is depth of focus. Thus a cat would have greater depth of focus horizontally, and less vertically, but only in daylight.

    • RobbertPatrison
      Posted August 11, 2015 at 11:18 am | Permalink

      You are right: The field of view is not dependent on the aperture, so there is no apparent optical advantage for a goat’s slit eyes. Also, I question how much the difference between horizontal and vertical daylight depth-of-focus would aid a cat in its depth perception. The lack of rigour in the optical analysis makes the entire premise of the article very questionable.

      • Posted August 11, 2015 at 11:37 am | Permalink

        I question how much the difference between horizontal and vertical daylight depth-of-focus would aid a cat in its depth perception.

        Something like that can be quite significant, especially as it pertains to out-of-focus areas. See here for an idea of what it can do with cameras:


        Now, consider that the cat’s eye’s aperture is predictably asymmetrical and in constant motion, and it’s easy to see how the brain can integrate that information to provide enhanced spatial resolution.

        …and, come to think of it…you know how a cat’s pupils get really wide just before the cat pounces? That’s going to simultaneously brighten the image and narrow the depth of field….


  28. TJR
    Posted August 11, 2015 at 4:52 am | Permalink

    Agreed that the lack of independence is a severe problem if the analysis doesn’t account for it. The CIs and p-values will be flawed. In effect, assuming independence makes the sample size look bigger than it really is, and hence the CIs narrower than they should be. Plus the calculations will in effect be assuming that the data are a random sample of species, whatever that might mean.

    They probably need some sort of hierarchical model, like a split-plot design or similar.

    However, in the chosen species the signal looks pretty strong, so it seems unlikely that they are just over-interpreting noise.

    Interesting stuff, anyway.

  29. Posted August 11, 2015 at 9:52 am | Permalink

    Reblogged this on librerianatura and commented:
    Un bellissimo articolo che spiega la forma delle pupille

  30. Posted August 11, 2015 at 11:24 am | Permalink

    Dennett is right – eyes are certainly a “good trick”.

  31. Ann German
    Posted August 11, 2015 at 1:38 pm | Permalink

    ms. trivia here, but I loved the opening that “here’s a picture . . . from National Public Radio.” Our descendants will read this and wonder how to watch a picture on the radio . . . ???

    • Posted August 11, 2015 at 2:34 pm | Permalink

      Actually, our descendants will likely be astonished to learn that there was once a content delivery medium that was restricted to nothing but audio. I suspect that, by then, “radio” will mostly be a technical term used to refer to a portion of the electromagnetic spectrum and only historical footnotes will mention today’s (yesterday’s) meaning. “Video killed the radio star” will be largely incomprehensible, with people wondering how a flat moving picture could cause the death of a pulsar.


%d bloggers like this: