Oribatid mite coprolite sightings – a transient craze ?
deutsche Version

Petrified wood does not only preserve the tissue structure of wood from bygone times but occasionally also its pests. Small wonder that irregular-shaped tiny holes in the wood with dark clots inside are interpreted as mites’ burrows with coprolites. Leaving the question aside whether the clots in some samples are really what they are believed to be, a very few observations make that interpretation highly dubious.
Nevertheless the oribatid mites, or rather their coprolites, became increasingly popular among palaeontologists in the 1990s. I did not try to find out when and where it started. The first dubious coprolite images which came to my notice were Fig.1 and Fig.2.
Clots misinterpreted as coprolites, Permian, Wetterau 
Fig.1: Silicified wood from the Wetterau area (Permian), Germany; cross section with local damage thought to be caused by mites in [1]. A few intact wood cells with clot inside (above right) raise doubts concerning the coprolite hypothesis.  Drawing after photograph in [1].
 Clots misinterpreted as coprolites, Permian, Schallodenbach

Fig.2 (right): Silicified wood from Schallodenbach (Lower Permian), Germany; longitudinal section with local damage thought to be caused by mites in [2]. The individual small clot deep inside a tracheid raises doubts concerning the coprolite hypothesis. Detail of photograph in [2].

Close inspection of the pictures raises doubts as one can see individual clots within intact wood cells where they hardly could have been dropped by mites. The doubts concerning mite frass are confirmed by another sample (Fig.3), where clots are seen in three cells in a row. Accepting the coprolite hypothesis would imply that three tiny mites had crept along neighbouring tracheids and deposited their droppings at exactly the same height.
No coprolites but wood rot, Permian, Schallodenbach
Fig.3: Silicified wood from Schallodenbach (Lower Permian), Germany, cross section with local damage. Three intact wood cells in a row with clot inside may serve as evidence against the coprolite hypothesis.
Own sample, kindly provided by Ch. Krüger , Schallodenbach.

In addition to the more or less rounded clots formed in the xylem cells, the sample of Fig.3 contains also distinctly angular clots formed in the larger pith cells (Fig. 4). Clots with angular shape should serve as convincing evidence against the interpretation as coprolites.

Clots from pith cell fills, Permian, SchallodenbachFig.4 (right): Silicified wood from Schallodenbach (Lower Permian), Germany, same sample as Fig.3, cross section of pith cavity in coniferous-type stem with only those pith cells preserved in outline which are filled with dark matter.

Nevertheless, the idea of discovering mite coprolites in one's fossil plant samples seemed to be so tempting to some palaeontologists that it spread worldwide, even to Antarctica [3]. Apparently not even the fact that oribatid mite fossils were absent in the Carboniferous, Permian, and Triassic where the alleged coprolites were most abundant [4] could dampen the eagerness with which the idea was accepted without being checked against common sense. Hence, sightings of oribatid mite coprolites were soon reported from samples stored at the Naturkunde-Museum Chemnitz [5-8]. About a century ago, the angular clots within plant tissue, including those in Fig.5, had probably been noticed by the clever J.T. Sterzel
(1841-1914) who was cautious enough not to propose an interpretation.
Clots misinterpreted as coprolites, Permian, Chemnitz
Fig.5: Angular clots in the tissue of the Permian climbing fern Ankyropteris brongniartii, recently interpreted as mite coprolites [5-7].
Sample Nr. K 4568, stored at the Museum für Naturkunde Chemnitz since 
Sterzel's time.

In view of the fossil evidence the following can be stated: Clots confined to the interior of intact cells (Figs.1-4) and angular clots with sizes and shapes like the cells of the damaged tissue harbouring them (Figs.4,5) cannot be interpreted as coprolites. Although the tell-tale shape and size is more or less evident from pictures in several publications, the authors apparently did not consider this worth mentioning.
The pictures most probably show a kind of wood rot caused by a fungus as explained in Rhynie Chert News 28.
From 2007 on when fossil oribatid mite coprolites were still a favourite topic among palaeontologists, this alternative interpretation had been sent to the authors listed below. It is hoped that such spread of information, together with the message conveyed by this essay, will help to make the craze of oribatid mite coprolite sightings, as well as other misconceptions, fade away and vanish.
See also the special contribution "Antarctic shit".

H.-J. Weiss      2010

[1] K. Goth, V. Wilde :  Fraßspuren in permischen Hölzern aus der Wetterau,
    Senckenbergiana letaea 72(1992), 1-6.
[2]  R. Noll, V. Wilde :  Conifers from the „Uplands“ – Petrified wood from Central Germany, 
    in: U. Dernbach, W.D. Tidwell : Secrets of Petrified Plants, D'ORO Publ., 2002.
[3] D.W. Kellog, E.L. Taylor : Evidence of oribatid mite detrivory in Antarctica during the Late Paleozoic and Mesozoic,
    J. of Paleontology 78(2004), 1146-53.
[4] C.C.Labandeira, T.L. Phillips, R.A. Norton : Oribatid mites and the decomposition of plant tissues in paleozoic coal swamp forests,
    Palaios 12(1997), 319-53.
[5] R. Rössler : The late palaeozoic tree fern Psaronius  -  an ecosystem unto itself,
    Rev. Palaeobot. Palyn. 108(2000), 55-74.
[6] R. Rössler : Der versteinerte Wald von Chemnitz, 2001, p 141,155,169.
[7] R. Rössler : Between precious inheritance and immediate experience,
    in: U. Dernbach, W.D. Tidwell : Secrets of Petrified Plants, D'ORO Publ., 2002.
[8] R. Rössler: Two remarkable Permian petrified forests,
    Geol. Soc. London Special Publ. 265(2006), 39-63.
quartz crystal with wood inside
Fossil Wood News  3

Site map
Fossil Wood News
Silicified wood