Tadpoles: super-abundant and super-significant in some aquatic environments. These ranid frog tadpoles were photographed in Germany by Markus Bühler.

Thanks to that recent Tet Zoo article about American spadefoot toads and their tadpoles, I’ve had visions in my mind of drying ephemeral pools in hot, arid environments, crammed with crowded, gasping tadpoles. All are racing to complete metamorphosis before what remains of the water is finally gone... It’s a sad scene, replayed annually in thousands of warm, shallow puddles and ponds across the hotter regions of the world.

Modern impressions made by tadpoles, photographed in a drainage ditch in Tennessee in 1958, from Maher (1962). A pipe serves as a scale.

But let’s not think about that too much. Before the end comes – before, that is, tadpoles either complete metamorphosis and leave the water, or die a horrible death that involves desiccation – tadpoles sometimes do an interesting thing. Well, they do lots of interesting things, but on this occasion I’m interested in one thing in particular: the tadpoles of some anuran species create polygonal or rounded hollows on the sediment surface of the pools they inhabit, thereby leaving a semi-permanent trace of their presence long after they’re gone. These structures – termed tadpole nests (Maher 1962, Cameron & Estes 1971) but said by some to be better referred to as tadpole holes (Dionne 1969) – can be remarkably regular and geometrical in appearance, as you can see from the photos here.

Photo montage (from Cameron & Estes 1971) showing how clusters of tadpoles (photographed in New York state) form polygonal impression on a sedimentary surface. Note that groups of tadpoles create each polygon - it isn't one tadpole per polygon.

Given that these structures form on the surface of aquatic sediments, they presumably have a propensity to be infilled by fine-grained muds and hence get preserved in the sedimentary record. And, indeed, a reasonable number of ancient impressions that seem to be ‘fossil tadpole nests’ have been reported from the geological record, many of which look just about identical to modern tadpole nests (Cameron & Estes 1971).

Triassic Batrachoides traces, thought by Hitchcock and others to be ancient tadpole traces. Image from Kindle (1914).

However, sedimentary structures that form in several other ways also strongly resemble tadpole nests, most notably interference ripple marks. Several classic exchanges in the literature – most notably those between pioneering ichnologist Edward Hitchcock and Charles Shepard during the 1850s and 60s – centred over whether given impressions were tadpole-generated or wave-generated. The structures described by Hitchcock, discovered in Massachusetts, were named Batrachoides*. Things then went quiet until the 1960s and 70s when the structures became the topic of debate again. Vialov (1964) argued that all so-called tadpole nests then reported had been misinterpreted and were either sedimentary artefacts resulting from the growth of algae across a sedimentary surface, or (in the case of Hitchcock’s Batrachoides) the result of wave interference.

* Not to be confused with Batrachoides, the extant toadfishes.

This debate is further complicated by the fact that modern depressions formed by tadpoles actually determine the position of mudcracks once the pools concerned have dried up (Metz 1986).

Metz (1986) showed how the formation of mudcracks was strongly associated with the distribution of hollows in mud formed by tadpoles. In the photo here (from Metz 1986), primary cracks ('P') run across the centre of the hollows, and both bifurcation cracks ('Be', 'Bo' and 'Bf') and quadriradiate cracks ('Q') are located in the centres of hollows.

The real identity of Hitchcock’s tadpole impressions was never definitely resolved, but they were apparently Triassic in age. This leads us to an interesting issue...

Is the record of ‘fossil tadpole nests’ at all informative as goes the nature and timing of the anuran fossil record, or is there some kind of woeful disjoint? Fossil tadpoles are known, most of which are Cenozoic, the oldest of which are from the Middle Jurassic (Roček et al. 2006). Hitchcock’s Triassic Batrachoides are, therefore, problematic. While there are some stem-anurans (like Triadobatrachus from Madagascar) from the Triassic, there aren't any reliable indications that Triassic stem-anurans await discovery in the fossil record of eastern North America [UPDATE: this is a somewhat controversial statement; see comments].

Rounded impressions, suggested to be the products of tadpole behaviour, from the Jurassic of Simla Himalaya, India. Image from Bhargava (1972).

Moving on, among the oldest possible fossil tadpole holes are from the Upper Jurassic Tal Formation of India (Bhargava 1972). These are subcircular traces approximately 10 mm across. They might be traces left behind by tadpoles, but they might not, since the similarity isn’t (in my opinion) especially strong. Also of interest is that these Indian impressions come from tidal sediments deposited on a beach – a habitat not typically considered a friendly one for tadpoles.

Poster-frog for quasi-marine habits in anurans: Fejervarya cancrivora, one of several extant frogs that will use sea-edge habitats and even swim in the sea on occasion. Photo by W. A. Djatmiko, CC BY-SA 3.0.

However, as Bhargava (1972) noted, tadpoles of several modern frog and toad species are in fact sometimes encountered in brackish and even saline water. Yes, that’s right, that ‘amphibians only occur in freshwater’ thing you’ve heard is not completely true – there are several frogs and toads that frequent seaside habitats, breed in brackish water, and even swim in the sea on occasion. The most famous of these is the so-called Crab-eating frog Fejervarya cancrivora of tropical Asia, but it isn't the only one.

Anyway, all this evaluation of Triassic and Jurassic ‘tadpole’ traces might be moot if these ancient impressions are inorganic in origin anyway. Cameron & Estes (1971) reviewed the record of supposed tadpole nests and argued that all fossil occurrences reported to date were of inorganic origin, and nothing to do with tadpoles or even vertebrates at all. This receives support from the fact that, beyond the Triassic, there are even alleged Paleozoic tadpole nests, the oldest of which are from the Devonian (Cameron & Estes 1971).

Fossil record or no, tadpole nests or holes or whatever remain neat and interesting, and I hope you enjoyed this brief review of what we know. Oh – as usual, there are scarcely any good, freely available images of these structures online. If you have any that you can share, let me know!

One last tadpole image before we go: this is a Western spadefoot (Spea hammondii) tadpole of the sort discussed in the previous article, photographed in a vernal pool. Image by Frank Landis, used with permission.

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

Refs - -

Bhargava, O. N. 1972. A note on structures resembling molds of tadpole nests in the Upper Jurassic Tal Formation, Simla Himalaya, India. Journal of Sedimentary Petrology 42, 236-245.

Cameron, B. & Estes, R. 1971. Fossil recent “tadpole nests”: a discussion. Journal of Sedimentary Petrology 41, 171-178.

Dionne, J. C. 1969. Tadpole holes: a true biogenic sedimentary structure. Journal of Sedimentary Petrology 39, 358-360.

Kindle, E. M. 1914. An inquiry into the origin of “Batrachoides the Antiquor” of the Lockport Dolomite of New York. Geological Magazine 1, 158-161.

Maher, S. W. 1962. Primary structures produced by tadpoles. Journal of Sedimentary Petrology 32, 138-139.

Metz, R. 1986. Control of mudcrack patterns by recent tadpole nests. Northeastern Geology 8, 1-3.

Roček, Z., Böttcher, R. & Wassersug. 2006. Gigantism in tadpoles of the Neogene frog Palaeobatrachus. Paleobiology 32, 666-675.

Vialov, O. S. 1964. Network structures similar to those made by tadpoles. Journal of Sedimentary Petrology 34, 664-666.