Beautiful painting of Wallace's flying frog (Rhacophorus nigropalmatus) by the unique Carel Brest van Kampen.

Episode 2 of David Attenborough’s Conquest of the Skies was on TV the other day, and I watched it (I livetweeted throughout, mostly because I wanted to talk about their portrayal of pterosaurs and Mesozoic theropods). And hence I have rhacophorid frogs on my mind – the mostly tropical Afro-Asian group that includes the famous Rhacophorus flying frogs, the best known member of which is Wallace’s flying frog R. nigropalmatus from Indonesia, Thailand and adjacent countries. As usual, flying frog were used by Sir David to help illustrate the diversity of animals that have evolved a gliding ability.

Rhacophorids are sometimes called flying frogs, shrub frogs, bush frogs, moss frogs, Old World treefrogs, or Afro-Asian treefrogs, and occur across sub-Saharan Africa, China, much of tropical Asia, Japan, the Philippines and Sulawesi. About 380 species are recognised as of early 2015. The last time I wrote about this group – December 2008 – this number was more like 290, so the rate at which new species are discovered and named is pretty impressive.

Cladogram from Meegaskumbura et al. (2002), the Sri Lankan taxa being shown in blue. Note the huge number of species that had not been named when this study was published.

A nice illustration of this is provided by Meegaskumbura et al.’s (2002) documentation of more than 100 new rhacophorid species on Sri Lanka alone (just 18 Sri Lankan rhacophorid species were known prior to their work), a discovery that makes Sri Lanka on par with Madagascar, New Guinea and Borneo in terms of anuran diversity.

And, yes, more than 100 new species announced in a single paper. If we look at the discovery record of various of the rhacophorid lineages, we see that – for example – 43 new species of Raorchestes, 30 new species of Rhacophorus, and 51 new species of Pseudophilautus have been named since 2000... 9 new Raorchestes species were named in 2014 alone (Frost 2014). As should be well known, the number of recognised amphibian species has sky-rocketed in recent years, and this really is because of newly discovered species, not just the result of splitting, taxonomic elevation of subspecies, or the recognition of cryptic species that can only be distinguished genetically.

The most famous illustration of a rhacophorid ever published: a gliding Rhacophorus nigropalmatus from Alfred Russell Wallace's 1869 The Malay Archipelago. Wallace wrote about this species and illustrated it in his notes (he didn't discover it though - that honour goes to Charles Hose).

The great paradox is that amphibians are in chronic global decline at the same time, and many species can no longer be located at all. Despite Meegaskumbura et al.’s (2002) 100+ new rhacophorid species, they were unable to find many that had been described in the 19th century, a discovery which implies that the species concerned have gone extinct. As you should also know, amphibian species are currently being ‘lost’ on a regular and worrying basis – we don’t talk of a ‘global amphibian crisis’ for nothing.

Most rhacophorids are arboreal or semi-arboreal, living in shrubs, trees and bushes from close to ground level to way up in the forest canopy. There’s some uncertainty over how high up in the canopy these frogs actually occur. Nature documentaries (like the aforementioned Attenbourough-led projects) create the impression that they really live tens of metres up in the high canopy but this is hard to confirm and has been doubted on occasion. Recently, however, individuals of some species (like Rhacophorus belalongensis on Borneo, named in 2008) have been recorded from tree-tops 10 m high. Members of some groups are associated with primary forests, but others inhabit agricultural fields, roadsides, cleared forest and villages. [Images below by Σ64 and Alpsdake.]

It has to be said that some rhacophorids are not really all that remarkable when compared to, say, familiar ranid frogs. This montage shows tadpoles and an adult of Rhacophorus arboreus. However, this species is a foam nester, on which see below. Tadpole photo at lower left by Σ64, other photos by Alpsdake. All CC BY-SA 3.0.

Such reproductive diversity

Specialised reproductive strategies are widespread across these frogs, and some of the techniques they use mean that they don’t have to come down to the ground to breed. Some (like some Philautus species) stick their eggs to the undersides of leaves above the ground and some Philautus species (like P. mjobergi) have been reported to be nepenthiphilous – that is, to lay their eggs inside pitcher plants. While some frogs definitely are nepenthiphilous, the only alleged rhacophorid eggs discovered inside a pitcher plant and subjected to molecular testing turned out to be from the microhylid species Microhyla borneensis (Hertwig et al. 2012). [Photo below by Katja Rembold.]

Frogs belonging to several lineages are documented as users of pitcher plants (this photo shows a Heterixalus inside a dead Nepenthes madagascariensis. Heterixalus is a hyperoliid, not a rhacophorid). It has been claimed that rhacophorids of several species use pitchers in this way, but the cases are either controversial or turned out to be erroneous. Photo by Katja Rembold, CC BY-SA 3.0.

Those Philautus eggs, by the way, don’t produce free-swimming tadpoles: Philautus species are direct developers, which means that the embryos change to froglets within the eggs, a free-living tadpole phase being absent (the developing embryos are lecithotrophic or endotrophic, meaning that they depend on a yolk store). Direct development is also the case in Pseudophilautus and Raorchestes.

While (as just mentioned) some of these direct developers stick their eggs to leaves that are alive and well above ground-level, others come down to the ground and lay their eggs beneath dead leaves. Meegaskumbara et al. (2007) said that these ground-breeding species “deposit their eggs in nests excavated on the forest floor” (p. 9). Waitaminute – frogs excavating nests? Really? I have to look into this...

The foam-nesters

Rhacophorid foam nests in Asia and Africa. At left: nests of Rhacophorus arboreus in Japan, photo by Alpsdake, CC BY 3.0. At right: Chiromantis rufescens foam nest in Gabon, photo by Brian Gratwicke, CC BY 2.0

Then there are those rhacophorids that manufacture arboreal foam nests [adjacent nest photos by Alpsdake and Brian Gratwicke]. The females exude a secretion that they (and their male partners) whip up with their legs to form a foam clump that’s attached to leaves, branches or aerial roots. It seems that the production of this secretion is quite costly and that a female needs to take a break and re-hydrate herself by soaking up standing water before she can complete a single nest.

This strategy is present in the Afro-Asian foam-nest frogs Chiromantis, most Rhacophorus species, and members of Polypedates. In some species – most famously the Grey foam nest treefrog C. xerampelina of south-eastern Africa – large numbers of these frogs sometimes choose to nest in the same place, meaning that branches or aerial roots can be festooned with whole lines of dripping foam nests. Actually, it isn’t just that the frogs ‘choose’ to nest in the same place – males will deliberately get in on the action if they see a pair working to make a nest, and the result is that single egg clutches are invariably fertilised by more than one male. Byrne & Whiting (2011) showed that this multiple paternity – technically, it’s simultaneous polyandry – assists in the survival of the resulting offspring, so it’s certainly in the interests of females to solicit as much male attention as possible during these breeding events. [Photos below by Brian Gratwicke, Kapenta and Chintan Sheth.]

Montage of Chiromantis and kin. At left: C rufescens (photo by Brian Gratwicke, CC BY 2.0). Top right: C. xerampelina (image by Kapenta, CC BY-SA 4.0). Lower right: Feihyla vittata (photo by Chintan Sheth, in public domain). The Feihyla species were once included within Chiromantis but have since been recovered in several alternative placements in molecular phylogenies.

The outsides of these foam nests dry to form a hard crust, thereby protecting the eggs within. However, monkeys, snakes and other predators will break into the nests and eat the eggs if they can. Most surprisingly, Fornasini’s spiny reed frog Afrixalus fornasini (a member of Hyperoliidae) is a documented foam nest predator, though it can only eat from the nest before the foam has dried (Channing 2001).

Polypedates leucomystax pair in amplexus. I think I spot some subtle sexual dimorphism. Image taken in Java, Indonesia, by W. A. Djatmiko, CC BY-SA 3.0

The eggs hatch inside the clump, the tadpoles dropping into the stream or pool (sometimes originally formed by rhinos or pigs) below after several days. Polypedates leucomystax bucks the trend by sometimes making foam nests on the ground (Inger & Stuebing 2005). [Adjacent photo by W.A. Djatmiko]. It seems that foam-nesting evolved just once within rhacophorids, since all foam-nesters belong to a single clade (Frost et al. 2006, Grosjean et al. 2008, Pyron & Wiens 2011).

Foam-nester tadpoles are ectotrophic: free-swimming and completing development outside the egg, and often with a schooling habit. Some live in muddy pools and are of typical, non-specialised morphology. Others (like those of Rhacophorus penanorum) are specialised stream-dwellers with streamlined bodies, sucker-like mouths and elongate, muscular tails. These tadpoles are rheophilous (associated with fast-flowing streams) and inhabit rocky pools that are sometimes also home to Megophrys/Xenophrys spadefoot tadpoles and Ansonia toad tadpoles (Haas et al. 2012).

A really interesting thing that’s been noted for rheophilous tadpoles is that their limb development seems to be offset, time-wise, relative to the condition in related, non-rheophilous species. This is presumably an evolutionary response to the fact that developing hindlimbs might affect their streamlining and ability to cling to rocks in fast-flowing water. They also keep sucker-like mouths and other features for longer than do other tadpoles (Nodzenski & Inger 1990). Accordingly, it can be difficult to say reliable things about their age and estimated metamorphic stage.

Amazing 'vampire tadpoles' of Rhacophorus vampyrus. (A) Schematic view of tadpole as seen from the front, (B) photo of the real thing. Image from Vassilieva et al. (2013).

Finally, there are yet other rhacophorids where egg clumps are laid in arboreal settings, but not in foam nests. In some Theloderma species, egg clumps are laid in water-filled tree hollows, and the tadpoles complete their development here. In captivity, these frogs will lay their egg clumps attached to the bark, just above the hollow, the hatching tadpoles then dropping into the water (Tapley 2009). Oh, there are also a few foam-nesting Rhacophorus species that lay their eggs in tree hollows, the most famous of which is R. vampyrus from Vietnam (after hatching inside a foam nest, the tadpoles drop into a water-filled tree hollow). This species saw international stardom a couple of years ago when it was revealed that the tadpoles have black, hooked fangs on the lower jaw that –it's presumed – are used when feeding on unfertilised eggs provided by their mother: these tadpoles, it seems, practise obligate oophagy, eating R. vampyrus eggs and nothing else (Vassilieva et al. 2013).

Holotype specimen of Philautus maia, showing eggs preserved in contact with ventral surface. Did this frog really carry its eggs around like this? Illustration from Günther (1876).

Finally finally, the possibility exists that a completely unique reproductive strategy was present in a species that now seems to be extinct. The holotype female specimen of Philautus maia, collected on Sri Lanka prior to 1876, had a disc-shaped mass of eggs attached to its belly, raising the possibility that members of this species carried their eggs around with them (Günther 1876). Alas, Meegaskumbura et al. (2007) discussed how unlikely this proposal was, concluding that a more plausible possibility is that the individual concerned was collected while in the process of laying and positioning her egg clutch on a leaf. Alas, only further observations can establish the ‘truth’ and... sadly, P. maia is one of those many, many frog species that seems to have become extinct between its 19th century discovery and the present. Don’t forget: Global. Amphibian. Crisis.

Mossy-skinned, warty-skinned, smooth-skinned, and with winglets

A moss-textured rhacophorid frog. Specimens of Vietnamese mossy frog or Tonkin bug-eyed Frog (Theloderma corticale). Top photo by Steven G. Johnson, CC BY-SA 3.0. Lower image by Václav Gvoždík, CC BY-SA 2.5.

External appearance is variable in rhacophorids. For all their fame as ‘flying frogs’, it has to be said that the vast majority look – to those unenlightened in anuran diversity – like standard ‘treefrogs’. They’re generally small (SVLs of 40 mm or less), wide-headed, big-eyed frogs with expanded digit-tips and a (normally) prominent tympanum. Many are smooth-skinned but spiny tubercles cover the skin in some taxa, and others are notably warty, with a rough, bumpy skin that aids camouflage.

The latter is most developed in the grotesque Rough treefrog Theloderma horridum of Thailand, peninsula Malaysia and Borneo. Indeed, this is one of several species (most of which belong to Theloderma) that resemble moss or bark in external texture and colour [adjacent photos of T. corticale by Steven G. Johnson and Václav Gvoždík]. T. asperum – patterned in brownish and pale blotches – superficially resembles a bird dropping and is sometimes called the Bird poop frog. Vocal sacs are absent in some taxa (like Nyctixalus) but big and obvious in others.

1854 illustration of Reinwardt's flying frog Rhacophorus reinwardtii by Jean Gabriel Prêtre. Image in public domain.

Wallace’s flying frog and the other gliding Rhacophorus species are pretty remarkable. They’re very big compared to most other members of the group, SVLs being 90-100 mm in females and 80-90 mm in males. Their fully webbed hands and feet are enormous, but they also have flaps of skin – winglets, if you like – on the arms and legs, and sometimes on the body. Glides of more than 15 m have been recorded. Exactly how many Rhacophorus species are true gliders is uncertain: the ability is confirmed for just a handful of species but more may have it (Inger & Stuebing 2005). Several smaller ones (including R. angulirostris, R. cyanopunctatus and R. gauni) have only partial digital webbing and either lack those skin flaps or only have small versions.

Finally, where do rhacophorids fit within the anuran radiation? Molecular studies find them to be close to Ranidae, the familiar neobatrachian clade that includes European water frogs, brown frogs, the American bullfrog, leopard frogs and so many others (Frost et al. 2006, Pyron & Wiens 2011). They’re clearly not at all close to hylid treefrogs (hylids are part of the same clade as glassfrogs, toads and kin). I also need to say that a huge amount of work – scarcely any of which is cited in the article you’re reading now – has recently been devoted to the in-group relationships of Rhacophoridae, several conventional ‘genera’ being the subject of substantial disagreement due to proposals that they might be paraphyletic or polyphyletic. At the risk of elaborating further, I must stop here. Oh, I seem to have blogged about anurans again.

The anuran takeover continues. Spot the rhacophorid! Illustration by Darren Naish.

For previous Tet Zoo articles on frogs and toads, see...

Refs - -

Byrne, P. G. & Whiting, M. J. 2011. Effects of simultaneous polyandry on offspring fitness in an African tree frog. Behavioral Ecology 22, 385-391.

Channing, A. 2001. Amphibians of Central and Southern Africa. Cornell University Press, Ithaca and London.

Frost, D. R. 2014. Amphibian Species of the World: an Online Reference. Version 6.0 (30th December 2014). Electronic Database accessible at http://research.amnh.org/herpetology/amphibia/index.html. American Museum of Natural History, New York, USA.

- ., 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.

Grosjean, S., Delorme, M., Dubois, A. & Ohler, A. 2008. Evolution of reproduction in the Rhacophoridae (Amphibia, Anura). Journal of Zoological Systematics and Evolutionary Research 462, 169-176.

Günther, A. 1876. Note on the mode of propagation of some Ceylonese tree-frogs, with description of two new species. Annals and Magazine of Natural History (4) 17, 377-380.

Haas, A., Hertwig, S. T., Krings, W., Braskamp, E., Dehling, J. M., Min, P. Y., Jankowski, A., Schweizer, M. & Das, I. 2012. Description of three Rhacophorus tadpoles (Lissamphibia: Anura: Rhacophoridae) from Sarawak, Malaysia (Borneo). Zootaxa 3328, 1-19.

Hertwig, S. T., Lilje, K. E., Min, P. Y., Haas, A. & Das, I. 2012. Molecular evidence for direct development in the rhacophorid frog, Philautus acutus (Rhacophoridae, Anura) from Borneo. The Raffles Bulletin of Zoology 60, 559-567.

Inger, R. F. & Stuebing, R. B. 2005. A Field Guide of the Frogs of Borneo. Natural History Publications (Borneo), Kota Kinabalu.

Meegaskumbura, M., Bossuyt, F., Pethiyagoda, R., Manamendra-Arachchi, K., Bahir, M., Milinkovitch, M. C. & Schneider, C. J. 2002. Sri Lanka: an amphibian hotspot. Science 298, 379.

- ., Manamendra-Arachchi, K., Schneider, C. J. & Pethiyagoda, R. 2007. New species amongst Sri Lanka’s extinct shrub frogs (Amphibia: Rhacophoridae: Philautus). Zootaxa 1397, 1-15.

Nodzenski, E. & Inger, R. F. 1990. Uncoupling of related structural changes in metamorphosing torrent-dwelling tadpoles. Copeia 1990, 1047-1054.

Pyron, A. R. & Wiens, J. J. 2011. A large-scale phylogeny of Amphibia including over 2800 species, and a revised classification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution 61, 543-583.

Tapley, B. 2009. Aspects of captive husbandry of Taylor’s Bug-eyed Frog, Theloderma stellatum (Taylor, 1962). Herpetological Bulletin 108, 31-33.

Vassilieva, A., Galoyan, E. & Poyarkov, N. 2013. Rhacophorus vampyrus (Anura:Rhacophoridae) reproductive biology: a new type of oophagous tadpole in Asian treefrogs. Journal of Herpetology 47, 607-614.