Portal de Periódicos da CAPES

Comments on Retallack, G. J. 2011: Problematic megafossils in Cambrian Palaeosols of South Australia

2012; Wiley; Volume: 55; Issue: 4 Linguagem: Inglês

10.1111/j.1475-4983.2012.01173.x

1475-4983

J. B. Jago, James G. Gehlîng, John R. Paterson, Glenn A. Brock,

Geological and Geochemical Analysis

The purpose of this comment is to question the organic nature of the material described by Retallack (2011) as ‘problematic megafossils’ from the Moodlatana and Balcoracana Formations, Lake Frome Group of the Flinders Ranges, South Australia. By way of introduction, we note that there is a considerable history of structures described from sedimentary rocks as fossils that do not stand up to scrutiny (Hofmann 1971, 1992, 2005). The literature is replete with what Hofmann (1992) regarded as dubiofossils. The onus should always be on the author to demonstrate biological origins. Retallack (2011, p.1223) in discussing the need to name problematic fossils states that ‘Vendobiota were informally noted by Mawson (1938, p. 259) as ‘fossil impressions resembling brachiopod or bivalve form’, but formal description of five species by Sprigg (1947) was needed before their global distribution and importance as Ediacaran fossils could be appreciated (Fedonkin et al. 2007).’ This is misleading because Mawson’s complete sentence is ‘Fossil impressions resembling brachiopod or bivalve form, but probably merely impressions of clay galls’. Mawson was in fact describing a section through what would now be termed the Ediacaran Rawnsley Quartzite, from about 1.6 km north of Parachilna Creek and about 50 km to the south east of where Sprigg (1947) initially found the Ediacara biota near Ediacara in the Rawnsley Quartzite. The sedimentary rocks of the Lake Frome Group are of Cambrian Series 2, Stage 4 and Cambrian Series 3, Stage 5 (early late to early middle Cambrian) in age and include facies with recognisable body and trace fossils (Jago et al. 2010). These formations represent the later stages of sedimentation in what has been referred to as the Adelaide Geosyncline (Sprigg 1952; Thomson 1969, 1970), Adelaide Fold Belt (von der Borch 1980) or Adelaide Rift Complex (Veevers et al. 1997). Given that the Moodlatana Formation includes both marginal marine and terrestrial units, palaeosol facies such as those described by Retallack (2008) cannot be ruled out. However, claims of fossils within palaeosols require very close scrutiny if only because such claims are difficult to verify in palaeosols of any age. Rocks of the Australian continent preserve one of the longest and most complex records of weathering and regolith history of any continent because the Australian continent has not suffered significant glacial erosion since the Early Permian (Crowell and Frakes 1971), nor has the continent suffered significant orogenic activity in post-Permian eras. Consequently, the weathering profile can extend up to >100 m deep for pre-Mesozoic rocks (Anand 2005). In this context, distinguishing body fossils of lichens, fungi or even bacterial mats in Cambrian palaeosols would require exceptional evidence, given the paucity of evidence of fossils other than vascular plants in palaeosols of younger age. At the very least, we would expect to see numerous specimens of consistent morphology in the suite of material described by Retallack (2011) and lodged in the South Australian Museum. Diagnoses should enable the reader to distinguish the described taxa. The author should be able to demonstrate a clear distinction between the putative fossil and the sedimentary textures of the rock samples as well as the weathering textures in the sample outcrop. We would expect the author to demonstrate that the weathering textures were indeed derived exclusively from the purported palaeo-weathering effects implied by the claim of a palaeosol. The only reliable method of doing this in the context of Australia’s complex weathering history is to identify both the palaeosol facies and the fossils, in borehole samples from depths below the local weathering profile. Retallack (2011) states that the taxa he described as Prasinema and Farghera do indeed occur in boreholes, although no such examples are figured. Given that early Palaeozoic fossil thallophytes of nonmarine origin are very rare, we would have expected the author to provide the highest quality of evidence for terrestrial thallophytes from the Cambrian. Claims for early terrestrial megascopic organisms need to be treated with considerable scrutiny when they are described from early Palaeozoic rocks. Thus, the onus was on Retallack (2011) to demonstrate that the taxa described from the early to early middle Cambrian Lake Frome Group of South Australia are megascopic thallophytes, fungi or some symbiotic combination. Any such fossil taxa require unique specified sets of physical characters that can be found by any palaeontologist examining the ‘type’ collection as lodged by the author. Unfortunately, the material figured and described by Retallack (2011) meets none of the criteria for biogenicity. The structures described by Retallack (2011) come from the upper Moodlatana and lower Balcoracana Formations in Ten Mile Creek in the Flinders Ranges. At this level, both the Moodlatana and Balcoracana Formations have a shallow marine to marginal marine depositional environment (Gravestock 1995). Retallack (2011) interprets the described structures as fossils preserved within palaeosols. While we do not wish to comment at this stage on the validity or otherwise of the supposed palaeosols, if the objects described as fossils were indeed fossils, then the morphology should be repeated from object to object within reasonable limitations. The material described by Retallack (2011) is housed in the collections of the South Australian Museum. All catalogue numbers refer to that collection. We have inspected the specimens closely and have looked at and photographed them with both low- and high-angled light coming from different directions. We have attempted to relate the specimens to the photographs and sketches provided by Retallack (2011). The descriptions would have been easier to follow if at least some of the illustrations (e.g. Retallack 2011, text-figs 5, 6, 8 and 11) had been presented in colour, especially as it is different coloured sediments that outline some of the items described by Retallack (2011). P42257 (Retallack, 2011, text-figs 4D, 5B), a sawn slab designated as holotype, is essentially a random pattern of intricately mixed red clays and silts with pale grey siltstone; the entire rock is slightly calcareous (Fig. 1A, B). There is no repeatable pattern. The holotype comprises a ‘concertina shape’ (Retallack 2011, p. 1229) somewhere within a cut slab of irregular shapes of red siltstone within pale grey siltstone. No single shape is repeated in the slab. Irregular mottles of the red siltstone occur throughout the rock. A, cut section through a mottled, ferruginous sandstone and bleached wispy, silt layers (white), which includes the holotype of a putative fungal fossil ‘Prasinema gracile’ (P42257). B, sketch by Retallack (2011, text-fig. 4D) of mottled sandstone claimed as the holotype of a fungal fossil ‘Prasinema gracile’ (P42257), for which there is no clear description of which objects are claimed fossils. C, ‘Prasinema nodosum’ (P42340a) is a cluster of dark, ferruginous areas of silty sand with bleaching haloes. D, ‘Prasinema adunatum’, holotype, (P42313; Retallack, 2011, text-figs 4E, 8C) a very small rock fragment with a bedding remnant of bleached fine grained sandstone lamination within ferruginous silty sandstone. E, ‘Erytholus globosus’ (P42256) is a random shape in a broken section of a poorly sorted ferruginous, muddy sandstone with deformed laminae (parallel to dashed line). The bulge with a mud flake in a median position, coinciding with a glued crack in the rock (arrow), was reconstructed by Retallack (2011, text-fig. 4J) as a filled quilted body with an axial structure extending into an ‘axial thread’. F, two claimed specimens (P42255) of ‘Erytholus globosus’ that were the basis of speculative reconstructions by Retallack (2011, text-fig. 4G–I), being slight protuberances (arrows) in ferruginous silty sandstone broken across poorly defined lamination found throughout the specimen (dashed line). G, ‘Farghera’ (P42314; Retallack, 2011, text-fig. 8H) is a mottled fragment of weathered, silty sandstone with patches of bleached sediment defining a crudely radiating pattern of red-brown (iron oxide) stained sediment in continuity with the whole rock. H, sketch of ‘Farghera’ (P42314) by Retallack (2011, text-fig. 4A)– a claimed lichen fossil from a ‘palaeosol’ that is indistinguishable from any other weathering pattern in this facies of the Moodlatana Formation. All scale bars represent 1 cm. Retallack (2011, p. 1226) gives the following diagnosis of Prasinema, ‘Network of fine (<2 mm diameter) filamentous green-grey, sediment-filled, irregular tubes, radiating and decreasing in abundance downward from a sedimentary surface, clear grey-green reduction haloes around the filaments contrast with red sedimentary matrix; unbranched or branching at irregular intervals and angles, without distinct orders of branch thickness.’ We are unable to distinguish between the tubes and the reduction haloes. There is no up or down marked on the slab of the holotype of Prasinema gracile; hence, we are unable to determine the accuracy of this part of the diagnosis. If anything, most of the pale grey sediment in the tubes is parallel with bedding as indeed is shown in Retallack (2011, text-fig. 5B). The tubes simply represent different oxidation states of iron within the sediment. Retallack (2011) gives no clear description, nor is there any indication on the sawn slab, as to which objects are claimed as fossils. P42312 (Retallack, 2011, text-fig. 8E) is on a small (2 × 1 cm) piece of red haematitic siltstone, which contains irregular layers of an off-white siltstone. Rather than obliquely orientated (as illustrated in Retallack 2011, text-fig. 8E), these irregular layers are probably parallel with bedding as are many in the Moodlatana Formation. The size distribution diagrams shown by Retallack (2011, text-fig. 7A–D in which the species name is given as Prasinema gracilis rather than P. gracile) have no real meaning because, as noted earlier, no single shape is repeatable. The logical extension of the interpretation by Retallack (2011) is that every grey-green irregular siltstone area within the haematitic silts of any bed within the Moodlatana Formation represents some sort of organic activity. P42310 (Retallack 2011, text-figs 4F, 8B) is the holotype. It is simply a very small (<1 cm across) grey-green siltstone area set in a haematitic siltstone. There are several examples of the same feature even in the very small (<2 cm across) piece of rock that contains the holotype. They have a completely random shape from circular to rectangular. To suggest that they are of organic origin must be regarded as extremely speculative at best. The measurements given by Retallack (2011, p.1232) are meaningless because orientation and putative dimensions are not replicable. P42311 in Retallack (2011, text-fig. 8D) is called P. gracile. The small grey-green siltstone areas within P42311 are indistinguishable from the holotype of P. nodosum. The grey-green areas are slightly more calcareous than the surrounding siltstone. The rock essentially comprises a haematitic siltstone with small areas of grey-green siltstone. P42317. We are unable to relocate the view shown in Retallack (2011, text-fig. 8F). This specimen simply comprises small (c. 1 mm) pale siltstone areas within a haematitic siltstone. P42340 (Fig. 1C; Retallack 2011, text-fig. 5C). The rock comprises micaceous haematitic siltstone with a 2-cm area of darker more Fe-rich sediment with intricately intermixed slightly calcareous pale grey siltstone patches on one side of the darker area. Similar pale grey siltstone patches are found elsewhere in the rock. We suggest that the paler areas are simply areas of slightly more calcareous siltstone with Fe2+, rather than Fe3+, ions present throughout most of the specimen. P42313 of Retallack (2011, text-figs 4E, 8C) is the holotype. It is an elongated area of grey-green siltstone within haematitic siltstone (Fig. 1D). The holotype is a small weathered chip (15 mm long, 10 mm wide, 5 mm deep) such as is found in enormous quantities in the Flinders Ranges. The diagnosis (Retallack 2011, p. 1223) reads ‘Prasinema of stout (2 mm) filaments, with a striated appearance and irregular swellings and thinnings; unbranched and subhorizontal in orientation’. The so-called striations are in reality laminations. It is unclear as to what is meant by filaments. Hence, neither the filaments nor the striations exist on the holotype. Given the small size of the holotype, it cannot be determined whether the orientation of the holotype is subhorizontal as stated in the diagnosis. The patterns drawn in the sketches shown in Retallack (2011, text-figs 4G, I, J) are very difficult to see on the specimens (Fig. 1E, F). The pattern shown in the bottom circle (see Retallack 2011, text-fig. 5F), that is, the holotype specimen (P42255, Fig. 1F), is caused by what appear to be four elongate, small pieces of haematitic mudstone lying parallel with each other, that is, they are mud flakes lying parallel with bedding. We are unable to locate the ‘median vertical column’ or the ‘thin axial thread’ as described and illustrated by Retallack (2011, p. 1223, text-fig. 4G–J). We are unable to locate the view illustrated by Retallack (2011, text-fig. 5D, specimen P42255) on a cut surface about 13 cm long and a maximum of 5 cm wide. We could not locate what Retallack (2011) illustrates in his text-figure 4H (on a sawn surface of only 12.5 cm by 5 cm maximum length and width). P42256 (Fig. 1E; Retallack 2011, text-figs 4J, 5E) is simply a bulge on a fracture surface formed as the rock was split open. The item figured as fossil is lying at right angles to the bedding (outlined by the white line in Fig. 1E). The longitudinal line shown in the photograph (Retallack 2011, text-fig. 5E) and sketched in text-figure 4J appears to be the junction of a crack following an elongated mud flake and also running around the bulge. It does not extend into ‘a thin axial thread’ as described by Retallack (2011). The spiral thread shown at the bottom of text-figure 4J is at the position of a crack where the specimen has been glued together. This crack continues under the main bulge of the specimen but is not connected to it; this part of the crack is not visible on text-figure 5E of Retallack (2011). Retallack (2011, text-fig. 9) produced measurement diagrams of Erytholus globosus (stated to be 145 specimens in the text and 146 specimens on the title bar of the figure), including dimensions such as width, thickness, chamber thickness and numbers of chambers. We are unable to identify the exact nature of these features on the available illustrations and specimens. Comparisons (Retallack 2011, p. 1234) of Erytholus with Ediacara biota fossils, such as Ernietta and Pambikalbae, are inappropriate as even the most poorly preserved specimens of these taxa require no special pleading to be diagnosed as of organic origin, unlike the forms described by Retallack (2011). P42314 (Fig. 1G, H; Retallack 2011, text-figs 4A, 8H). In Farghera sp., the red siltstone that makes up Farghera is outlined by the surrounding lighter coloured siltstone; however, the red material continues on into the main part of the surrounding rock from which it cannot be distinguished. The pattern is produced by weathering and mobilisation of iron oxides on a broken fragment of silty, ferruginous sandstone. The pattern cannot be distinguished from any other pattern in this facies of the Moodlatana Formation. The specimen P42315 (Retallack 2011, text-figs 4B, 8I) is at right angles to bedding and is simply a random pattern of red siltstone and lighter coloured siltstone. We suggest that most, if not all, of the patterns described by Retallack as fossils were caused by a combination of bedding-related features and the modern weathering cycle. None of the ‘species’ described by Retallack show a repeatable pattern, and hence, we conclude that none of the material described as fossils by Retallack is of organic origin. Editor. Svend Stouge