A Devonian plant that would not fit in
deutsche Version

Fossil finds which do not seem to fit to known species are not uncommon and keep the palaeontologists wondering. Most often it is found out where they fit in after all, otherwise they may indicate the presence of something new. Really intriguing are the nematophytes, which have got names but have not yet got a branch on the Tree of Life. The Lower Devonian Rhynie chert has provided some, including new ones. The Rhynie chert is expected to yield more surprises, and here is another one which awaits interpretation
unknown plant section, Rhynie
Fig.1: Unknown fossil in Rhynie chert with granular aspect. Width of the picture 9.5mm.

Fig.2 (below): Detail of Fig.1. Width of the picture 1.5mm.

unknown plant section, Rhynieunknown plant section, Rhynie

Fig.3 (left): Angular grains of unknown origin, detail of Fig.1. Width of the picture 0.83mm.

The plant or plant parts presented here look like a bag filled with grains (Figs.1,2). Their partially straight contours could be crystal faces or replicas of plane cell walls. Apparently, silicification has produced quartz powder in addition to the compact grains, providing good optical contrast.
The dark grains show a slight tendency of being arranged in chains, as seen in Fig.3 at some place on the right, for example. This indicates that a process spreading from cell to cell may have affected the silification of the interior. Phenomena of such kind, perhaps mediated by some rot fungus, have produced the cell-size clots or angular grains repeatedly misinterpreted as mite

The grain size of 20...35Ám is smaller than the cell sizes of any of the 7 vascular plants hitherto discovered in the Rhynie chert, and also smaller than the "coprolites" found in petrified wood.
The black appearance is brought about by two effects: The transparent grains let the light pass into depth and therefore look much darker than the white powdery quartz, and some have black inclusions of decayed organic matter. Near the surface of the grain bag the dark grains are so crowded that one cannot see whether or not the plant had an epidermis.
cuticle (?) of unknown plant section, RhynieThere is a small spot of decayed and vanished tissue where the former surface is faintly indicated by a thin dark line (Fig.4). Such line is also seen in some places on the other half of the cut specimen. The thin lines along parts of the contour might be the remains of a cuticle on an epidermis. detail of unknown plant section, Rhynie

Fig.4 (left): Cuticle (?) seen as a dark line covering a hole under the surface (below left in Fig.1).

Fig.5 (right): Squeezed end of the bag in Fig.1 or end of stalk with holdfast ? Width of the picture 0.7mm.

There is one feature in Fig.1 which is either highly significant or a mere artefact due to partial decay and compression: The bag of more than 1mm thickness suddenly narrows into an indistinctly seen extension to the right, poorly silicified and traversed by cracks. It resembles a stalk with a (tripartite ?) foot or holdfast, width 0.6mm, at its end (Fig.5).
As the bag is also seen on the other side of the cut gap of 1mm or more,
and with the surfaces of the halves standing nearly perpendicular on the cut faces, one can conclude that what is seen in Fig.1 is a lengthwise section of a bag of little more than 1mm thickness and at least 3mm width. If the narrow extension really were a stalk, it would be ribbon-like.
The fact that no indication of vascular tissue is seen on either half of the divided bag does not mean there is none but it suggests that possibly there is none.
The absence of hyphae and tubes excludes any affiliation with fungi or nematophytes. The presence of a cuticle (Fig.4) does not necessarily indicate a land plant. Organisms living in an aquatic habitat which repeatedly falls dry may develop a cuticle, as it is known from nematophytes. The habitat which turned into the Rhynie chert by silicification harboured aquatic and subareal plants, fungi, and creatures. At a vertical distance of about 3cm below Fig.1, lots of squeezed and a few well preserved Rhynia sections are seen, together with several charophyte specimens resembling Palaeonitella, indicating subaerial and submerged conditions at times.
A plant with bag-like body, stipe, and holdfast suggests comparison with the recent Brown Alga Laminaria. The comparison leads to a bigger and a smaller problem: This fossil is more than twice as old as the allegedly first Brown Algae [1]. Hence, there must have been algae with the aspect of Brown Algae long before, which poses the problem of where to place them on the phylogenetic tree.
The fact that this alga had lived in the Devonian freshwater-dominated habitat at Rhynie does not preclude an interpretation as a precursor of Brown Algae since it is known that among the numerous extant Brown Alga species there is a small fraction of freshwater species [2]. This small alga could have been a freshwater Brown Alga of its time, which would be the only one found hitherto.
Although a final interpretation of the fossil has to be left to those who might find more specimens of this kind sometime or perhaps have a bright idea just now, a few tentative conclusions suggested by the observations have been presented here.

Sample: Rh22/1 (0.21kg), found by Sieglinde Weiss at Castlehill in 2009.

H.-J. Weiss      2013,    modified in 2021

[1]  Th. Silberfeld  et al.: A multi-locus time-calibrated phylogeny of the brown algae ... Molecular Phylogenetics and Evolution 56 (2010) 659–674.
[2]  J.D. Wehr: Freshwater Algae of North America, Chapter 19, Brown Algae, 2014, Acad. Press.

Site map
Rhynie Chert News
Rhynie chert