By Matthew Cobb
One of the great things about my job – I’m a university professor – is that in order to be able to keep up with my research and teaching, I need to surf the endless tide of scientific discovery, and try not to drown in it. Sometimes it all gets too much and I feel like I’d like to disconnect the internet and go and raise goats in the hills. But more often, I come across an absolute delight.
Today’s offering is this paper from earlier in the year – a study of fossilised millipede turds from the early Devonian (413-418 MY ago) – which I came across while preparing for tomorrow’s second year lecture on myriapods. The authors of the study – Dianne Edwards, Paul Selden and Lindsey Axe – looked at about 50 tiny fossils in rocks exposed by a stream north of Brown Clee Hill, in Shropshire, UK. These rocks had previously revealed evidence of wildfires, in which organic material was rapidly transformed into charcoal, and finally into exquisite fossils:
The enigmatic, tiny fossils – a little over a millimetre long – are of tiny blobs, either segmented, or single. By looking at the inside and the outside of the fossils, Edwards and her co-workers concluded that they were coprolites – fossilised turds, to you and me.
The minute remnants of these long-ago lunches revealed something amazing about the animal that excreted them. Inside the coprolites were fossilised sheets of cells and cuticle that were not derived from higher plants (that is, they are not the remains of spores, stems, sporangia or anything else beginning with ‘s’). Instead, the authors identify these remains as coming from extinct organisms called nematophytes which are mysterious but have “fungal affinities” (it was first thought that nematophytes were plants – the paleobiologists are arguing about this). Something – with a rather small anus – was eating these nematophytes over 400 million years ago.
So what made the turds?
The authors do not have any direct evidence for this – there is no defecating equivalent of a mortichnia, no beast has been caught in the act – but by comparing these coprolites with others, they conclude that they would have been made be early terrestrial invertebrates: mites, collembolans or millipedes. And because of their tiny size, they suggest that they came from millipedes.
Surprisingly – or not, depending on your point of view – they say ‘little is known about modern millipede fecal pellets’, beyond that they are approximately the same size, at least in some groups. However, Lynne Boddy at Cardiff is apparently carrying out an in vivo study of millipede poo, which suggests that millipede turds are ellipsoidal and when the animals have been eating fungal hyphae, these are found in the feces.
Apart from the amazing fact that we can infer the presence of arthropods from so long ago, the really interesting thing about this paper is that it forms part of a growing tendency to try and understand the ecology of long-lost landscapes. In this case, it appears that the millipedes were specialised on eating these nematophytes (all the remnants in the coprolites were of these fungal relatives).
This ecosystem sprang up shortly after arthropods invaded the land – the earliest fossil is a bit of a millipede, from around 450 million years ago, which isn’t too long after the earliest plant fossils found on land (460 million years). Prior to that, it appears probable that the land was covered either with a thin layer of algae, or, going much further back, with a biofilm of unicellular bacteria or archaea. The importance of understanding the ecosystem when multicellular life invaded the land is that it gives us a way of looking at the earliest forms of terrestrial carbon cycle – as things eat other things, they move and transform carbon. The following two figures show the order in which we currently have evidence for the invasion of the land by plants and animals – these are the latest possible dates; new fossils may well push the dates back earlier.
Appearance of various plant fossils, taken from Kenrick et al (2012). NB ‘nematophyte cuticle’ on the left
Appearance of major arthropod groups in the fossil record. The millipede turds are from the Lochkovian. Figure from Kenrick et al (2012)
These figures are taken from recent review by Paul Kenrick, Charles Wellmann (hi, Charles!), Harald Schneider and Gregory Edgecombe. This review, which is freely available, is a great description of our current understanding, which suggests that terrestrial life – and ecosystems – may have started even earlier than we thought.
But to really understand the ecosystem, you not only need fossils of the participants, you need some fossils of what the organisms are doing with each other, including crapping it out at the other end.
Paul Kenrick, Charles H. Wellman, Harald Schneider and Gregory D. Edgecombe (2012) A timeline for terrestrialization: consequences for the carbon cycle in the Palaeozoic. Phil. Trans. R. Soc. B : 519-536 (open access! Hooray)