About the SA Blog Network

Tetrapod Zoology

Tetrapod Zoology

Amphibians, reptiles, birds and mammals - living and extinct
Tetrapod Zoology Home

Glassfrogs: translucent skin, green bones, arm spines

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Hopefully you can see why they're called glassfrogs.

Glassfrogs, or centrolenids, are wide-skulled, long-limbed arboreal little frogs (SVL 20-60 mm), unique to the Central and South American cloud and rain forests. Not until 1951 did this group get recognised as a distinct and nameable entity: prior to this, species within the group (known to science since 1872) had been classified as part of Rhacophoridae, the Old World ‘bush frogs’ or ‘shrub frogs’. Most glassfrogs lay their eggs on vegetation overhanging water or on rocks above the water surface and their tadpoles live in nearby streams. Many texts refer to them as ‘glass frogs’; I here follow several recent publications (and the trend that’s occurring in biological nomenclatural in general) in referring to them as ‘glassfrogs’. [Adjacent photo from here.]

Humeri of several male glassfrogs, from Guayasamin et al. (2009). From top to bottom: Centrolene geckoideum, Ce. pipilatum, Chimerella mariaelenae, and Ikakogi tayrona. The spines are on the anterior edge of the humerus, close to the proximal end. Note the sometimes enormous ventral flanges present on the posterior margins.

Glassfrog eyes are set on the tops of their heads, they have adhesive disks on their digit tips, and – while they are generally greenish on their dorsal surface – they derive their common name from the fact that they lack pigment on their ventral surface, meaning that their undersides are essentially transparent. I have no idea why this is, and I’m not sure that anyone else does. Even stranger, many species have green bones (not all do: Ikakogi and the Hyalinobatrachium species have white ones).

The terminal phalanges of glassfrog digits are T-shaped (this is also the case in a few other neobatrachian groups, like poison-arrow frogs), the males of some species possess spines on their upper arms (these are used in territorial combat), and the two uniquely elongate ankle bones that characterise anurans (the tibiale and fibulare) are fused into a single element.

Centrolenidae is not a small or insignificant group: there are currently about 150 named species (a number that has increased substantial since the late 1980s; back then, there were about 65 recognised species), with numerous additional ones known but awaiting description (Cisneros-Heredia & McDiarmid 2006, pp. 12-13).

A revised taxonomy

Espadarana prosoblepon (note the humeral spine), photo by D. F. Cisneros Heredia, licensed under Creative Commons Attribution-Share Alike 2.5 Generic license.

Until recently, all glassfrogs were grouped into just three genera: Centrolene (distinctive due to its humeral spines), Hyalinobatrachium (distinctive due to its prominent, white liver), and Cochranella (which lacks both of these features). A fourth ‘genus’, Nymphargus, was named in 2007 for species previously included in Cochranella but lacking hand webbing (Cisneros-Heredia & McDiarmid 2007).

However, a phylogenetic analysis of molecular characters found the first of those three genera to be polyphyletic and Nymphargus to be paraphyletic to Centrolene (Guayasamin et al. 2008). A taxonomy created to reflect this phylogeny resurrected or coined seven additional genera: Chimerella, Espadarana, Rulyrana*, Sachatamia, Teratohyla, Vitreorana and Celsiella (Guayasamin et al. 2009). Centrolene in the strict sense (the clade that includes C. geckoideum – the very first glassfrog species to be named – and its close relatives) was found to be the sister-group to Nymphargus. Vitreorana, Teratohyla, Rulyrana, Sachatamia, Cochranella sensu stricto and Espadarana grouped together as a clade which formed the sister-group to a Centrolene + Nymphargus clade; this former clade was named Cochranellini. The few Celsiella species (previously lumped into Cochranella) formed a clade with Hyalinobatrachium. [Teratohyla image below by Bgv23.]

Mating Teratohyla spinosa pair. So translucent. Teratohyla species lack humeral spines. Speciation within Teratohyla seems to have been driven by the rising of the Andes. Image by Bgv23, licensed under Creative Commons Attribution 2.0 Generic license.

* This name was invented by combining the first letters of the names Ruiz-Carranza and Lynch with ‘rana’ (meaning frog). Pedro Ruiz-Carranza and John D. Lynch are amphibian specialists who have “contributed enormously to the understanding of centrolenid diversity, biology, and evolution” (Guayasamin et al. 2009, p. 36).

It seems that a divergence into a ((Centrolene + Nymphargus) + Cochranellini) clade and Celsiella + Hyalinobatrachium clade occurred early in glassfrog evolution: Guayasamin et al. (2009) named the first clade Centroleninae and the second Hyalinobatrachinae. While molecular and morphological characters have been recognised for both clades, Guayasamin et al. (2009) drew attention to the very different fighting behaviours of these frogs. Male hyalinobatrachines battle on top of leaves and wrestle, with their fighting poses often resembling amplexus. In contrast, male centrolenines hang from leaf edges by their hindfeet and wrestle while dangling in the air! Hyalinobatrachines lack the large and often formidable humeral spines present in some centrolenines. [Image below by Mauricio Rivera Correa and from the CalPhotos database.]

Cochranella savagei, image by Mauricio Rivera Correa, licensed under Creative Commons Attribution-Share Alike 2.5 Generic license.

A taxon previously included in Centrolene (C. tayrona) but found in Guayasamin et al.’s (2009) phylogeny to group outside the Hyalinobatrachinae + Centroleninae clade was given its own ‘genus’ – Ikakogi – by these authors. Accordingly, they hypothesised that Ikakogi represented a lineage at least as old as both hyalinobatrachines and centrolenines (incidentally, fossil centrolenids are unknown, and hypotheses about when the divergence events might have occurred depend on distribution and biogeography). However, Pyron & Wiens (2011) found Ikakogi to be part of Centroleninae (specifically, the sister-taxon to the rest of the clade). Whereas other glassfrog species are consistent in whether they prefer the upper or lower surfaces of leaves as egg-deposition sites, I. tayrona is unusual in that it will deposit eggs on either surface. It’s further unusual in that females guard the egg clutches. There are other glassfrog species where egg-guarding occurs, but (so far as we know), it’s always males that do the guarding.

I. tayrona also possesses those humeral spines. If Ikakogi is hypothesised to be the sister-taxon to all other centrolenines, it could be that the spines are a primitive character for glassfrogs lost early on in hyalinobatrachines and also in several centrolenine lineages. Notably, the humeral spines present in the various glassfrog lineages differ in where they’re placed on the edge of the humerus and often seem to be formed from different parts of the bone. This could mean that the spines have actually evolved independently on two, three or more separate occasions – that might seem remarkable but similar things are known to have happened elsewhere (e.g., ear asymmetry in owls has evolved on at least seven separate occasions). With Ikakogi interpreted as a centrolenine (Pyron & Wiens 2011), things are different, since humeral spines are then unique to this lineage. Still – did they evolve once, or several times within Centroleninae?

Where in the anuran tree?

Simplified hyloid phylogeny, based on Frost et al. (2006) but mostly consistent with the topology also recovered by Pyron & Wiens (2011). Bufonidae: image by Froggydarb, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license. Dendrobatidae: image by Cliff, licensed under Creative Commons Attribution 2.0 Generic license. Ceratophryidae: image by Grosscha, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license. Centrolenidae: image by Mauricio Rivera Correa, licensed under Creative Commons Attribution-Share Alike 2.5 Generic license. Leptodactylidae: image in public domain. Hylidae: image by Kropsoq, licensed under Creative Commons Attribution-Share Alike 3.0 Unported license.

It is uncontroversially accepted that glassfrogs are part of the major neobatrachian clade termed either Hyloides, Hyloidea or Bufonoidea: that is, they’re part of the same major group as treefrogs, toads, horned frogs and poison-dart frogs. Glassfrogs used to be considered especially close to hylids (treefrogs), but data from mitochondrial DNA indicates that they’re more closely related to toads, leptodactylids and kin (Darst & Cannatella 2004, Frost et al. 2006, Pyron & Wiens 2011) and are especially close to leptodactylids. Frost et al. (2006) proposed the name Diphyabatrachia for the centrolenid + leptodactylid clade.

It has also been suggested that glassfrogs are the sister-taxon of Allophryne ruthveni (e.g., Austin et al. 2002, Guayasamin et al. 2008, 2009), a controversial and problematical toothless hyloid that has often been given its own ‘family’, Allophrynidae (a second species of Allophryne has recently been discovered, but I don’t think it’s been published yet. Please let me know if you know otherwise [UPDATE: it’s A. resplendens Castroviejo-Fisher et al., 2012 (Castroviejo-Fisher et al. 2012). Thanks to Diego Cisneros-Heredia for this.]). Frost et al. (2006) even subsumed Allophryne into Centrolenidae and changed the taxonomy such that their version of Centrolenidae included an Allophryninae and a Centroleninae. Guayasamin et al. (2009) decided that the best course of action was to keep Allophryne outside of Centrolenidae but to use the name Allocentroleniae for the Allophryne + Centrolenidae clade. The simplified cladogram shown here doesn’t include the names Diphyabatrachia or Allocentroleniae (nor is Allophryne featured), but hopefully you get the point.

Lest we forget: the Global Amphibian Crisis

This article is a re-vamped version of a section of text that appeared on ver 2 in 2007. As you will know, I’m sure, frogs all around the world are currently beleaguered by various issues, one of the most worrying of which is the spread of the aquatic fungal pathogen Batrachochytrium dendrobatidis (Bd for short).

The Golden toad (Incilius periglenes) of Costa Rica, extinct c. 1989 and one of c. 200 anurans hypothesised to have become extinct or near-extinct due to climate change, pollution, pathogen spread, or a combination of some or all of these factors.

A huge amount of research on Bd has been done since it was first recognised as a cause of amphibian decline. We now know that several anuran species are immune (or largely immune) to the disease caused by the pathogen and that they act as reservoirs for the pathogen, passing it to other, more suspectible species by sharing the same habitats. Ironically, the African clawed frog Xenopus laevis and American bullfrog Lithobates catesbeianus (or Rana catesbeiana) – the two anuran species that people have taken with them all around the world – are among those Bd-resistant species. The spread of Bd has also been linked with climate change. This will all be very familiar stuff if you know anything already about anurans or conservation, but it remains an area of major concern and certainly hasn’t gone away. For news on anuran conservation, the spread of Bd and other, related issues, keep an eye on ZSL’s Frog Blog, Amphibian Ark and Frog Matters.

For previous Tet Zoo articles on anurans, see…

Refs – -

Castroviejo-Fisher, S., Pérez-Peña, P. E., Padial, J. M. & Guayasamin, J. M. 2012. A second species of the family Allophrynidae (Amphibia: Anura). American Museum Novitates 3739, 1-17.

Cisneros-Heredia, D. F. & McDiarmid, R. W. 2006. A new species of the genus Centrolene (Amphibia: Anura: Centrolenidae) from Ecuador with comments on the taxonomy and biogeography of Glassfrogs. Zootaxa 1244, 1-32.

- . & McDiarmid, R. W. 2007. Revision of the characters of Centrolenidae (Amphibia: Anura: Athesphatanura), with comments on its taxonomy and the description of new taxa of glassfrogs. Zootaxa 1572, 1-82.

Darst, C. R. & Cannatella, D. C. 2004. Novel relationships among hyloid frogs inferred from 12S and 16S mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 31, 462-475.

Frost, D. R., Grant, T., Faivovich, J., Bain, R. H., Haas, A., Haddad, C. F. B., De Sá, R. O., Channing, A., Wilkinson, M., Donnellan, S. C., Raxworthy, C. J., Campbell, J. A., Blotto, B. L., Moler, P., Drewes, R. C., Nussbaum, R. A., Lynch, J. D., Green, D. M. & Wheeler, W. C. 2006. The amphibian tree of life. Bulletin of the American Museum of Natural History 297, 1-370.

Guayasamin, J. M., S. Castroviejo-Fisher, J. Ayarzaguena, L. Trueb y C. Vilá. 2008. Phylogenetic relationships of glass frogs (Centrolenidae) based on mitochondrial and nuclear genes. Molecular Phylogenetics and Evolution 48, 574-595.

- ., S. Castroviejo-Fisher, L. Trueb, J. Ayarzagüena, M. Rada, C. Vilá. 2009. Phylogenetic systematics of Glassfrogs (Amphibia: Centrolenidae) and their sister taxon Allophryne ruthveni. Zootaxa 2100, 1-97.

Pyron, R. A. & Wiens, J. J. 2011 A large-scale phylogeny of Amphibia including over 2,800 species, and a revised classification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution 61, 543-583.

Darren Naish About the Author: Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at! Follow on Twitter @TetZoo.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 34 Comments

Add Comment
  1. 1. David Marjanović 1:10 pm 01/25/2013

    Teratohyla is “monster treefrog”, Vitreorana is literally “glass frog”.

    Rhacophoridae, the Old World ‘bush frogs’ or ‘shrub frogs’

    Including the “flying” frogs!

    Link to this
  2. 2. naishd 1:22 pm 01/25/2013

    Yes, I resisted the urge to explain the etymologies of all those new names. Little known fact: Hyla (universally regarded as feminine) was apparently named for Hylas, companion to Heracles and abductee of water nymphs.


    Link to this
  3. 3. dfcisneros 3:03 pm 01/25/2013

    Great article! Thanks for citing me and using one of my photos.

    Glassfrogs are such amazing frogs, but we still know so little about them. For example, in addition to being translucent, these frogs have the membranes (peritonea) covering their internal organs with a rather thick layer of guanine! You can see in the first photo of this article: a red heart but white stomach, liver and intestines. We do not know why they are translucent… but usually when they are, their organs are covered by this white layer of guanine.

    In the last part of your article you mention the amphibian crisis. Glassfrogs are among the amphibians that have declined since the 1980′s. Yet their declines do not seem to be related with the Bd desease. Most species of glassfrog are still present (in low numbers) and even when some are affected by BD, the impact is low. The biggest problem for glassfrogs populations are apparently habitat changes and local and regional climate changes (not necessarily linked with global warming), especially in the tropical Andes.

    By the way, the new species of Allophryne is already described, it was called Allophryne resplendens, you can find the original description here:

    Diego F. Cisneros-Heredia

    Link to this
  4. 4. souhjiro 5:21 pm 01/25/2013

    perhaps the skin transparency as adult is derived from their tadpoles being such specialized?(fossorial on streambeds ,relying on cutaneous breathing mainly)

    Link to this
  5. 5. naishd 5:25 pm 01/25/2013

    Diego – thanks so much for your comment :) (and I’m pleased that you’re ok with the use of your photo). Very interesting what you say about glassfrogs and Bd. And I’m pleased to hear about the second species of Allophryne – I recall seeing mention of it several years ago but was unaware of the paper until now.

    More hyloids to appear here soon!


    Link to this
  6. 6. bjnicholls 6:32 pm 01/25/2013

    It looks like the muscles must also be translucent, not just the skin. Perhaps the white guanine protects gut microbes from adverse light exposure.

    Link to this
  7. 7. Heteromeles 7:28 pm 01/25/2013

    Cool article, as usual.

    If one is intensely cynical (as I am), one might not regard the relative immunity of Xenopus and Lithobates as ironic, but as fairly predictable. Why would anyone cart them all over the place, unless they were tough, which tends to translate as highly disease resistant? How many other frog transports were tried, and failed when the animals died of disease? This is similar to the problem conservationists have with invasive plants. Many of them were chosen (and, in fact, bred) to be hardy and “easy to grow” under the profoundly punishing conditions of the average clueless homeowner’s garden. Small wonder that they take over any nearby disturbed space, given any chance at all.

    Link to this
  8. 8. Rajita 10:14 pm 01/25/2013

    Thank you for that bit if information on Guanine. Guanine is previously known from fish scales where it forms a peculiar crystalline biomaterial along with hypoxanthine which result in a metallic sheen. Apparently this type of guanine crystal seen in fish scales cannot be formed in the lab. It is that special crystal form that allows the dispersion of light that results in iridescence. This iridescence is believed to be adaptive for the fish but it is exact role is not clear to me. Some say it literally dazzles predators. But the guanine in the frogs looks more like the laboratory guanine crystals with a white color. Interesting that in two distinct groups of vertebrates guanine is accumulated thus.

    Link to this
  9. 9. David Marjanović 8:12 am 01/26/2013

    Yes, I resisted the urge to explain the etymologies of all those new names.

    Why? :-)

    Interesting that in two distinct groups of vertebrates guanine is accumulated thus.

    It’s also common outside of vertebrates; many animals use guanine crystals as a reflective layer in their eyes.

    And many squamates have a pigmented peritoneum (though that’s just ordinary melanin), interpreted as protection from the sun which apparently the inner organs need more than the skin and the muscles. ~:-|

    Link to this
  10. 10. SRPlant 8:55 am 01/26/2013

    “Yes, I resisted the urge to explain the etymologies of all those new names.

    Why? ”

    Indeed, you have us intrigued…

    Link to this
  11. 11. naishd 9:24 am 01/26/2013

    A huge number of Tet Zoo articles have histories that go like this: (1) Darren thinks “uh oh, nothing new on Tet Zoo for a while, need to put up something asap”, (2) Darren looks for short bit of text – frequently from Tet Zoo ver 2 – that can quickly be republished as a stand-alone bit of text, and successfully finds one, (3) Darren decides that said bit of text needs revision and update in view of additional or new stuff that should be said; text changes from c. 200 words to bloated >1000 word monster, takes hours to prepare rather than minutes (with those hours mostly involving the small ones of the morning when normal people are asleep). In cases such as this, I make a deliberate effort to avoid tangents that will involve the adding of several 100 words (or more).

    I produced this glassfrog article because I became frustrated by my inability to finish another one on hyloid anurans; that latter article was started because of stuff I said in the Tet Zoo 7th birthday article… it would be extremely frustrating if I failed to make the time to finish the glassfrog one. Hey, if someone paid me to blog full-time I’d be able to get through everthing :)


    Link to this
  12. 12. SRPlant 10:15 am 01/26/2013

    Explanation accepted!

    (I will now return to my afternoon nap)

    Link to this
  13. 13. naishd 10:22 am 01/26/2013

    You can have _naps_ in the afternoon? Wow :)


    Link to this
  14. 14. SRPlant 11:40 am 01/26/2013

    I spent the pm dreaming of abduction by water nymphs (all your fault!)

    Link to this
  15. 15. Heteromeles 12:42 pm 01/26/2013

    I guess the first question about glassfrog pigmentation is what tends to hunt them? Does being translucent help for camouflage? Or is it some sort of parasite defense (assuming parasites are negatively phototropic)?

    Personally, I keep hoping the reason they’re translucent is because they exhale oxygen, during the day…

    Link to this
  16. 16. Andreas Johansson 2:09 pm 01/26/2013

    You can have _naps_ in the afternoon?

    You can. I find that the usual result is that you then can’t fall asleep in the evening and end up typing stuff on the intertoobz in the small hours.

    Link to this
  17. 17. naishd 6:58 pm 01/27/2013

    Thanks to all for comments. Heteromeles (comment 7) said…

    “If one is intensely cynical (as I am), one might not regard the relative immunity of Xenopus and Lithobates as ironic, but as fairly predictable. Why would anyone cart them all over the place, unless they were tough, which tends to translate as highly disease resistant? How many other frog transports were tried, and failed when the animals died of disease?”

    Yeah, all good points – I think you’re partly right (that is, that species widely used as lab animals or transported around the world are robust compared to many others). But I think you might also be partly wrong – that is, the resistance to Bd of, and extensive use by people of, X. laevis and L. catesbianus may well have been mostly or wholly coincidental. X. laevis proved a useful laboratory/med science animal and hence got transported all over the world, but I don’t think anybody was trying to use other anuran species this way before they began to take advantage of X. laevis (they hadn’t discovered those species, nor learnt of their vulnerabilities). Similarly, the American bullfrog has been so extensively exploited due to its size and availability – it just so happens that it’s tough and Bd-resistant as well.


    Link to this
  18. 18. Heteromeles 12:56 am 01/28/2013

    @17: You may be right Darren. I don’t know how many frogs were tried and found wanting, and I tend to think that most species that are farmed or used as lab rats tend to be fairly resistant to begin with. So yes, it’s coincidence, but if I had to blindly guess which species would be resistant to a widespread pathogen, I’d predict anything that’s invasive and widely transported. By the way, any news on how cane toads do with Bd?

    Link to this
  19. 19. naishd 4:01 am 01/28/2013

    Cane toads and Bd… I’m sure you can guess. Yup, resistant. It seems that resistant species are – as a generalisation – lowland animals with heightened immune responses, so it makes sense that they’re the same ones with a high degree of physiological/behavioural ‘robustness’. Some frog populations in susceptible species have been said to have developed a degree of Bd resistance. I need to read up on that; last I heard, the populations concerned (e.g., of Mountain yellow-legged frog Rana muscosa) had simply been exposed to low-virulence Bd.


    Link to this
  20. 20. Dartian 5:52 am 01/28/2013

    Centrolenidae is not a small or insignificant group: there are currently about 150 named species

    How do centrolenids differ ecologically from hylids? This might just reflect my own ignorance, but – superficially at least – to me glassfrogs would seem to occupy a very similar niche to that of the ‘true’ treefrogs. What makes it possible for these two species-rich anuran groups to coexist in the Neotropics?

    Link to this
  21. 21. Heteromeles 5:27 pm 01/28/2013

    Hmmm. So living at the downstream end of the watershed makes one more resistant, possibly more invasive? That’s good to know.

    Personally, I’m trying *not* to use this as a general metaphor for the utility of grad school, tempting though it may be.

    Link to this
  22. 22. Heteromeles 11:15 pm 01/28/2013

    Hate to double-post, but I wonder whether an advantage of “glassiness” is in camouflage. Obviously, the frog isn’t transparent. However, the point of camouflage is to break up the shape. Having the organs somewhat visible inside helps to break up the outline of the frog, since the shape and position of the organs will change depending on the perspective it is viewed from, thereby disrupting the shape of the frog.

    Link to this
  23. 23. naishd 3:54 am 01/29/2013

    Thanks for comments – no time to try and provide any answers right now, but I’ll deal with some of the issues raised here in another glassfrog article, to appear soon…


    Link to this
  24. 24. David Marjanović 3:47 pm 01/29/2013

    I wonder whether an advantage of “glassiness” is in camouflage

    The glassy side, bizarrely, is the one you don’t see unless you flip the frog over.

    Link to this
  25. 25. David Marjanović 3:48 pm 01/29/2013

    …or it’s particularly bloated, like the female Teratohyla spinosa shown above.

    Link to this
  26. 26. Heteromeles 4:52 pm 01/29/2013

    @24: Actually David, I get that. However, if you look at the frogs in the pictures, you’ll note that you can still see the interior. I’m playing with translucency as a form of camouflage, on the assumption that predators search for frogs that are opaque. The shapes of the viscera and bones could help break up the frogs’ overall outline.

    Link to this
  27. 27. Jerzy v. 3.0. 7:58 am 01/30/2013

    Small pieces of glass are difficult to see in water, so maybe the glass-like quality is a camouflage blending with wet surfaces?

    Maybe frog-eating snakes and other predators see in polarized light, and get confused by glassfrogs on rain-covered plants?

    Link to this
  28. 28. David Marjanović 9:59 am 01/30/2013

    True, there’s some translucency, and that would suddenly explain the green bones.

    (But then, gars have green bones, too, and they’re extremely intransparent.)

    Maybe frog-eating snakes and other predators see in polarized light

    Most of them are vertebrates, so probably not.

    Link to this
  29. 29. Heteromeles 1:05 pm 01/30/2013

    I just finished reading a book called Sleights of Mind, about the neurobiology of magic. The central point of the book (sorry for the spoiler) is that magical illusions are not exactly failures of human perception. Instead, human brains do a *lot* of post-processing to give us our perception of reality, and magic exploits the way we process data to make tricks successfully. According to the book, for example, human optic nerves only pass the equivalent of a few megapixels of data (about the range of a low-end phone camera), but our brain is constantly stitching together multiple pictures, sound data, memory, and prediction, to give us the illusion we see more than we do. The result of having this system is that we can be fooled by illusions that exploit with the expectations that help our brains function.

    It’s a fun book to read. I’d like to suggest it’s also a rich research ground for what I might call “magical ethology.” Simply put, how do animals respond to illusions?

    The point here is not that an animal can’t see physically something (as with a frog hypothetically invisible in polarized light), but that it can’t mentally see something (for example, it sees frog outlines on the assumptions that frogs are opaque, and misses the outlines of translucent frogs because it processes them as debris).

    You can have a lot of fun seeing what human illusions animals do and don’t respond to. To pick one example, most people know the phantom ball trick with dogs: you pretend to throw a ball, and the dog tracks the motion and chases the non-existent ball (humans fall for this illusion too). I’ve noticed that breeds such as weimaraners are more susceptible to phantom balls than bearded collies are. What I’ve also noticed is that birds don’t respond to it at all. I’ve tried this with pigeons, blackbirds, and cowbirds. If you throw a scrap of food from your lunch, they’re all over it. Pretend to throw a scrap of food, and they don’t even blink. Presumably this is because they see faster (sample their eyes more times per second) and have a much shorter persistence of vision effect, so they actually can track the food leaving my fingers, while humans and dogs have to guess and get fooled.

    I don’t want to derail this thread too badly, but the point is that it’s not just how a predator processes light in the eyeball, it’s how the animal processes the data its optic nerve feeds to its brain. I suspect that there are a large number of experiments researchers could run to see how this works. For instance, can a snake see a frog it can’t smell? What shape does it require to strike at a frog accurately, assuming it can smell the frog? Human vision can be primed by sounds, so why should we assume that snake vision is totally independent of olfaction?

    Link to this
  30. 30. Jerzy v. 3.0. 5:18 pm 01/30/2013

    Birds are well known to see polarized light. Maybe water reflections confuse them differently.

    Perception tricks must be common in animal world. Certainly, many animals use perception tricks at humans. I remember catching oystercatcher chicks for ringing. Little squeaking blighters were impossible to catch by hand, because a running chick, when you try to grab it, changes direction and you grab the air near it. The only thing which worked was to swipe an arm along the ground to knock the zigzagging chick off its feet.

    Another example is mosquito. Did you notice how difficult is to swat a mosquito (I am talking about European ones)? Because it is designed to fool the supposedly most intelligent species on the planet.

    Link to this
  31. 31. llewelly 8:04 pm 01/30/2013

    Maybe transparent skin is a side effect of some other beneficial attribute; perhaps glassfrogs benefit from thin skin and the resulting easier absorption of oxygen and water , and transparency is merely a side effect of thin skin.

    Link to this
  32. 32. AlexanderBerg 8:35 pm 01/30/2013

    [from Darren: sorry, delayed by spam filter]

    Now how about the Rhacophorid species Rhacophorus dulitensis that seem to share characters with the neotropical glass frogs? Might this shed some light on the puzzle?

    Here’s some pictures I took on Borneo of this (here and here), there are more on the net if you search for it too.

    (More “herps” and other critters if anyone is interested)
    Although they are not as transparent as glass frogs, they are still transparent compared to other Rhacophorus species. They are even slightly transparent dorsally as well, as can be seen in the photographs. Not only that, they have visibly green bones, and what I guess is guanine (from reading above) in their abdomen. The guanine surrounded stomach is better shown in some of the pictures that come up on google.

    In the case of R. dulitensis it is not likely to camouflage in an aquatic setting, as they do not even enter the water as adults when breeding, instead making a foam nest on leaves above small ponds.

    Maybe its to reduce the shadow-silhoutte on the backside of leaves that they sit on? It would not make it dissapear, just not be as pronounced compared to an opaque frog – so its not really a good explenation :-/

    Link to this
  33. 33. llewelly 9:17 pm 01/30/2013

    I have to say I find the polarized light discussion interesting but puzzling. Animals known to use polarized light to confuse predators, such as certain cephalopods, do not necessarily have transparent skin.

    The kind of polarization that results from transparent surfaces is strongest when the transparent surface is planar; the angle the light strikes the transparent surface at affects the angle of the resulting polarization, so a curved transparent surface (such as the skin of a frog) or worse, a bumpy transparent surface will result in photons with many different polarizations, and weaker correlation in their polarizations. It’s the correlation in polarization angles across many photons that defines strength of polarization, so the polarization resulting from a curved transparent surface will usually be weaker than that resulting from a planar transparent surface. So transparent skin is not necessarily a good way to produce polarization.

    The case for glassfrog skin being able to produce any useful polarization effect seems to poor.

    But if the polarizing skin speculation is nonetheless granted, I find it puzzling that insects have not yet been mentioned. Many insects detect polarized light well. I don’t know of any insect predators of glassfrogs, but they are small enough for it be possible, and they live in areas where large insects occur. Polarized vision among vertebrates, especially tetrapods, seems to be relatively rare.

    Link to this
  34. 34. Heteromeles 11:58 pm 01/30/2013

    @Jerzy: I’m not surprised about the oystercatcher chicks dodging. Birds see things fast. A long time ago I had a pet pigeon (broken winged bird, kept inside). That bird was a real brat, and at one point I threw a balled-up sock at him to get him to leave something alone. He simply stood still and watched it pass an inch from his head. With his tiny little head and apparent lack of depth perception, he’d still been able to figure out that the sock would miss him. I knew something was on target because he’d dodge before it got there.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article