The evolution of cat coat patterns

Why are some species of kittehs plain, while others have spots, stripes, or more elaborate patterns? A provisional answer comes from a new paper by William Allen et al., “Why the leopard got his spots: relating pattern development to ecology in felids”, in the Proceedings of the Royal Society.  The paper’s title, of course, comes from Rudyard Kipling’s Just So Stories.  And the short answer is this: the coats of wild cats help camouflage them, and what pattern evolves depends on where the species lives.

The simple answer comes from a rather elaborate analysis.  The authors set up the paper with what I think is a good specimen of clear scientific writing.  It’s not Joyce, of course, but these guys know how to write. I love the alliteration of “flanks of felids” and the breeziness of “pounce or quick rush.”

The patterns displayed on the flanks of felids are intriguing in their variety. Previous studies of the adaptive function of cat coat patterns have indicated that they are likely to be for camouflage rather than communication or physiological reasons [1,2]. The primary hunting strategy of felids is to stalk prey until they are close enough to capture them with a pounce or quick rush [3,4]. As hunts are more successful when an attack is initiated from shorter distances [5,6], cats benefit from remaining undetected for as long as possible and camouflage helps achieve this. Many smaller cats are also likely to be camouflaged for protection from predation [7].

The authors first note that others before them have suggested—and supported with some data—the idea that spotted or stripey cats live in forested habitats, and plain cats in open habitats.  But they quantify this “complexity” by doing a developmental analysis of coat patterns on pictures taken from the internet.  I won’t go into the details, but they match the photographs with patterns generated from a mathematical model in which pattern results from the interaction of two diffusible chemicals along gradients of the body.  Given a model that matches an existing pattern (they used 35 species of felids), they could then encompass “pattern” in the mathematical constants involved in generating it.  They could then correlate these constants with various aspect of cat ecology: where they live, preferred times of activity, how big they are, what they eat, and how social they are.

Here’s an example of a cat that came out “plain” in their analysis: the caracal (Caracal caracal), from Africa and the Middle East:

Nine of the 35 species were considered “plain.” Here’s a cat considered “patterned and complex”: the gorgeous clouded leopard (Neofelis nebulosa), from southeast Asia:

Sixteen species were considered patterned, with four of these, including the clouded leopard, as “always complex.”  The other ten were considered “variable”,” since there was polymorphism: individuals within a species can look quite different.

The results?

  • Pattern itself, whether complex or not, was significantly associated with habitat, with more patterned cats in more “closed” habitats (forest, jungle, etc.).  Plain cats are found in open habitats like grasslands, deserts, and mountains.
  • More irregular patterns, like the cloud leopard, are significantly associated with tropical forests and other “closed” environments.
  • “The more time cats spent in trees, the more likely they were to be patterned.”
  • Pattern polymorphism, as in the melanism of “black panthers,” was significantly associated with living in temperate forests that vary seasonally and also with habitat generalism. This supports the idea that “disruptive selection,” that is, selection for different patterns in different places, maintains the intra-specific variation in coat color.
  • There were a few “outliers,” or exceptions—cats that had patterns not fitting into the habitat correlations given above.  One is the very rare bay cat (Catopuma badia; I’ve posted on it before), which is plain though it lives in tropical rainforest:

And another outlier is the black footed cat of Africa (Felis nigripes), which is patterned though it lives in open habitat (savannah, grassland, and semi-desert):

The authors note that the tiger is the only wild cat with vertical stripes, and the common notion that this camouflages them in grassland is unfounded: tigers don’t live in grasslands.

The conclusion, then, is that the patterns of cat coats reflect, in large degree, selection for camouflage in their natural habitats. This camouflage almost certainly evolved to hide them from prey, and, in smaller cats, predators as well.

I love the inclusion of a Kipling quote in their conclusion (reference “45” is to the Just So Stories):

These findings support the hypothesis that felid flank patterns function as background matching camouflage. Evolution has generally paired plain cats with relatively uniformly coloured, textured and illuminated environments, and patterned cats with environments ‘full of trees and bushes and stripy, speckly, patchy-blatchy shadows’ [45].

Now the sample size—35 species—is not large, and some of the associations were barely significant from a statistical standpoint.  This could reflect the low power of tests in small samples. Nevertheless, the study offers a good working hypothesis for the evolution of pattern not just in cats, but other species that “need” to be cryptic.  What remains to understand are those outliers like the bay cat, and also the existence of developmental change of pattern, in which some species are patterned when young and lose the patterns when they get older.  Lions, which are spotted as cubs, are a good example of this:

This change might not be adaptive per se, but simply be an atavism: a holdover from an ancestral spotted pattern that still persists in the young.

_________

Allen, W. L., I. C. Cuthill, N. E. Scott-Samuel and R. Baddeley. 2010.  Why the leopard got his spots: relating pattern development to ecology in felids.  Proc. Roy. Soc. B online: doi: 10.1098/rspb.2010.1734

29 Comments

  1. Sajanas
    Posted October 27, 2010 at 8:47 am | Permalink

    Could it be that lions don’t need the patterns as adults, but it is useful for cubs, since predation could still affect them?

    • Posted October 27, 2010 at 12:00 pm | Permalink

      I don’t think that would make sense.

      Both adults and juveniles live in the same environment. If spots would help juveniles avoid detection and thus predation, then similar spots would also help the adults avoid detection and thus be more effective predators themselves.

      If I had to make a WAG, it would be that the lion’s ancestors were all marked similarly to the cubs and lived in an environment where such markings were effective camouflage. The environment changed (or the lions moved) to one better suited for plain markings.

      The adult coats changed because of the selective pressures for those markings. The juvenile coats didn’t change because nothing is stupid enough to attempt to launch a sneak attack on a lion cub; the fact that they’re a bit less camouflaged is irrelevant.

      Getting even more speculative, it might even be helpful for them to be a bit more visible, the same way that poisonous insects advertise their presence. That cute little spotted ball of fluff over there? Run away now, because mama’s certainly nearby and has probably already started her attack run.

      Cheers,

      b&

      • Dominic
        Posted October 28, 2010 at 2:32 am | Permalink

        I wonder how the cave lion was patterned as a cub as in the few contemporary human drawings of them the adults have no spots. And the extinct cats like the ‘American Lion’ which is now considered a relative of the jaguar I believe. Fascinating.

      • darrylzero
        Posted November 24, 2010 at 5:58 pm | Permalink

        I’d disagree with that. Just because one camouflage works for adults, that doesn’t mean it would work for cubs. My basic point is that when something is prey, it needs to avoid predators. Predators have better depth perception, eyes are in front of their head to judge distance. Conversely, a predator hiding from prey needs to hid from something that has more like motion detector style eyes. Bigger field of vision, less depth perception. Something with a paler less patterned coat may be less noticeable to those types of eyes. My basic point is that not everything sees or perceives visual patterns the same. There is a chance that because of this, they have different patterns at different life cycle points.

  2. Sal Bro
    Posted October 27, 2010 at 9:03 am | Permalink

    I wouldn’t mind seeing a similar analysis of house cats. For example, I think my kitteh’s ancestors may have spent a good deal of time on unfinished wood floors. (Gratuitous kitteh plug.)

    • Sal Bro
      Posted October 27, 2010 at 9:21 am | Permalink

      Actually, I have a question. This study is suggesting an adaptation of coat color to environment in wild felids, but I’ve wondered whether domesticated felids (whose coat color is more variable and, presumably, no longer linked to their environment) show any preference for areas where their coats are more camouflaged.

      • Dominic
        Posted October 28, 2010 at 3:03 am | Permalink

        There is the well know effect under domestication, of humans selecting for less aggressive animals, which produces adult animals that are more like the young of the wild equivalent so dogs often have floppy ears. Belyaev did this with wild foxes selecting for non-agression, & a by-product seems to be bi-coloured coats, like cats, dogs, horse, cows – even domestic fowls & pigeons show that. Ask a geneticist why the genes for those traits are linked. A good book about this is Stephen Budiansky’s book ‘The Covenant of the Wild’.

    • Kingasaurus
      Posted October 27, 2010 at 10:03 am | Permalink

      Jerry, count me in as someone fascinated about how this information would apply to domestic cats.

      Our vet told us once that the complex nature of domestic coat color/pattern genetics and expression was extremely difficult to detangle.

      You can basically throw up your hands when trying to predict what a given litter is going to look like.

      Why are domestic cats’ color/pattern so varied?

      • Posted October 27, 2010 at 10:24 am | Permalink

        I actually have no idea, but one guess is that predation pressures for domestic cats are very low, and so there can be lots of variation in markings since such traits would be neutral. Their wild ancestors would have faced much stronger selection pressures on colour and markings.

        Cuteness and friendliness might be selected for instead — who can resist a cute kitteh?

      • Sal Bro
        Posted October 27, 2010 at 10:49 am | Permalink

        Why are domestic cats’ color/pattern so varied?

        Well, various coat colors have been selected for by humans. In addition, though, we’ve also indirectly selected for different coat characteristics as a byproduct of breeding for desirable behavioral attributes–see, for example, the domesticated silver fox.

      • ritebrother
        Posted October 27, 2010 at 11:38 am | Permalink

        From what little I’ve read, I recall that the “mackerel tabby” coat patern is sort of the ground state when coats are not selected for. That is, solid colors, for example, are quicklly lost in a few generations of interbreeding, and the mackerel tabby pattern becomes the default.

        • ritebrother
          Posted October 27, 2010 at 11:38 am | Permalink

          I meant selcted for ‘artificially’ (i.e. selective breeding).

        • Posted October 27, 2010 at 12:30 pm | Permalink

          I’ve done a bit of Web research inspired by your post. The Mackerel Tabby coat pattern seems to be, from what I can gather, the same as that of the African Wildcat (Felis silvestris lybica), which shares a very recent (~10,000 years ago) ancestor with the domesticated housecat. The Egyptians invited some wildcats to come in from the…er…heat. Modern housecats are descended from those ancient Egyptian deities; their cousins can still be found in the wild. And not only can successfully interbreed with them, but are hard to differentiate from them…this would be one of those cases where the dividing line between species is very blurry.

          And, you know what? Baihu’s markings are indistinguishable, to me, from some of the pictures on the ‘Net of African Wildcats. See:

          http://news.bbc.co.uk/2/hi/science/nature/3611453.stm

          http://www.animals-pictures-dictionary.com/rec/122-African-Wildcat-Felis-silvestris-libyca/

          Damn, but I love that cat. Learning how ancient and noble his heritage is and how close his ties to that heritage are just makes me love him even more.

          Cheers,

          b&

        • Thornavis.
          Posted October 27, 2010 at 12:38 pm | Permalink

          That would make sense as I understand the domestic cat is considered to be a mixture of the African and European Wildcat, both of which are striped.

        • Dominic
          Posted October 28, 2010 at 3:06 am | Permalink

          ‘Tortoiseshell’ – & low & behold, a Wikipedia pageon cat coat colouring!

          http://en.wikipedia.org/wiki/Cat_coat_genetics

    • neil
      Posted October 27, 2010 at 1:37 pm | Permalink

      One coat-selection story I have read about F. catus (although I do not know whether it is true–sounds a bit like an urban legend) concerns the inevitable white mark on black cats. Apparently the all-caring, god-loving, morality-fount church of the middle ages burned black cats alive. However, a white mark was considered a “god mark” and the cat was spared.

      • Mike from Ottawa
        Posted October 29, 2010 at 9:50 am | Permalink

        “Apparently the all-caring, god-loving, morality-fount church of the middle ages burned black cats alive.”

        Except in those places where that self-same church didn’t. Sorry, that’s just a fact. Didn’t mean to harsh your buzz.

  3. Dominic
    Posted October 27, 2010 at 9:22 am | Permalink

    Ever see a King Cheetah? Spots become stripes. Interesting as there is so little genetic variety in the cheetah.

    http://en.wikipedia.org/wiki/King_cheetah#King_cheetah

    • Posted October 27, 2010 at 9:37 am | Permalink

      Wow…you can’t see it on any of the pictures below, but Baihu has a strikingly similar trio of stripes down his spine. Some of his dorsal stripes also turn into cheetah-like spots; you can see some of those right above his right rear leg in the first picture.

      Cheers,

      b&

  4. Posted October 27, 2010 at 9:30 am | Permalink

    The authors note that the common idea that tigers have vertical stripes (and are the only cat with such a pattern) because they live in grassland is unfounded: tigers don’t live in grasslands. (Emphasis added.)

    Erm…I have proof right in front of me that tigers are not the only cats with vertical stripes. Well, he was in front of me, until he squirmed out of my lap.

    Behold Baihu:

    Bonus points for an explanation of his name….

    I’ve always assumed he’s best described as an American Shorthair tabby. Anybody got a better classification for him? He was born to a feral mother in Tempe, Arizona.

    Cheers,

    b&

    • Dominic
      Posted October 28, 2010 at 3:18 am | Permalink

      Vey handsome cat – & for striped relatives of Baihu, the link I think I put up the other day to the BBC website where they capture Felis sivestris kittens on film in the Scottish highlands –

      http://news.bbc.co.uk/earth/hi/earth_news/newsid_9085000/9085644.stm

      I would bet that under true ‘wild’ conditions in the countryside (as opposed to urban ‘wild’ conditions) feral cats are most likely to revert to the North African wildcat ‘type’ like Baihu.

  5. Buford
    Posted October 27, 2010 at 10:16 am | Permalink

    Plain cats plainly come mainly from the plains?

    • Kevin
      Posted October 27, 2010 at 11:21 am | Permalink

      Where it rains…but mainly in Spain.

  6. Posted October 27, 2010 at 11:29 am | Permalink

    Lovely shout-out to the Just So Stories – I grew up on that book. I used to beg my mother to read ‘The Elephant’s Child’ to me for years after I knew how to read – I wanted to hear it. It occurs to me now that that was my introduction to an awareness of literary style.

  7. Posted October 27, 2010 at 12:03 pm | Permalink

    And another outlier is the black footed cat of Africa (Felis nigripes), which is patterned though it lives in open habitat (savannah, grassland, and semi-desert):

    The black footed cat appears juvenile in form. As I understand it, evolutionary pressures can arrest development in order to overlay new development patterns. As with the lion cubs, perhaps the ancestors of black footed cats were patterned as juveniles and grew “plain” as adults, and the “arrested” descendents now are patterned.

    • Thornavis.
      Posted October 27, 2010 at 12:42 pm | Permalink

      Wouldn’t that rather call into question the anti predator hypothesis as applied to small cats ?

      • Posted October 28, 2010 at 12:04 am | Permalink

        Perhaps. But if the trade off has greater survival value (invokes a suite of previously neutral mutations), then offspring can live with the patterned coat. Hard to say without genetic analysis. Tks.

        • Posted November 2, 2010 at 8:12 am | Permalink

          How about the following: The environment of cubs and small cats (an adult black-footed cat is only 2kgs) at their scale is certainly not ‘open’. Even though a semi-arid environment may seem open to a 1.5m tall human, it is not ‘open’ to a 20cm tall cub or cat. Perhaps the spotted cubs & small cats are better camouflaged among the grasses and low shrubs than if they had a plain coat.

          Lion cubs are not exempt from predation. They can be the victims of infanticide if another male lion takes over a pride, and other adult carnivores such as leopard and hyena may well kill lion cubs to reduce competition.

  8. Hempenstein
    Posted October 27, 2010 at 3:12 pm | Permalink

    Maybe the outliers are rare (you said Catopuma badia is; is Felis nigripes rare too?) because they have the wrong pattern for their environment?


3 Trackbacks/Pingbacks

  1. [...] sharing because it’s interesting science. Jerry Coyne’s Why Evolution Is True blog has this fantastic article about some research that has been done on cat coat patterns. I thought I’d share it and hope [...]

  2. [...] and Cats: Jerry Coyne talks about the evolution of patterns on cats (e. g., spots, smooth coats, etc.) Evidently, much of it was adaptation to their environment (e. [...]

  3. [...] recently posted about a new analysis of cat coat color patterns by William Allen and colleagues from the University [...]

Follow

Get every new post delivered to your Inbox.

Join 29,398 other followers

%d bloggers like this: