<<

PALAEONTOLOGICAL HERITAGE STUDY: COMBINED DESKTOP AND FIELD-BASED ASSESSMENT

Rehabilitation of National Route (Section 3, km 24.2 to km 75) between Cradock and ,

John E. Almond PhD (Cantab.) Natura Viva cc, PO Box 12410 Mill Street, Cape Town 8010, RSA [email protected]

February 2013

1. SUMMARY

The South African National Roads Agency Limited (SANRAL) is proposing to rehabilitate Section 3 of the National Route R61 (km 24.2 to km 75) between Cradock and Tarkastad, Eastern Cape. The project involves widening of the roadway and of all stormwater structures along the route. Road material is to be sourced from five new or existing borrow pits and one hard rock quarry. A Phase 1 palaeontological heritage assessment for the road project has been commissioned by Arcus GIBB (Pty) Ltd in accordance with the requirements of the National Heritage Resources Act (Act 25 of 1999).

Section 3 of the R61 traverses the outcrop area of continental sedimentary rocks of the Upper (Tarkastad Subgroup, Supergroup) of Early to Middle age. These are cut and baked by numerous dolerite intrusions of the Karoo Dolerite Suite of Early age. Towards Cradock (Graaff-Reinet and Middelburg 1: 250 000 sheet areas) the sedimentary bedrocks belong to the -dominated that was deposited in arid braided fluvial settings following the catastrophic end- mass event. Further east towards Tarkastad (Queenstown and King William’s Town 1: 250 000 sheet areas) the sedimentary bedrocks are assigned to the slightly younger Burgersdorp Formation comprising recessive- reddish and braided river channel . Both these rock successions are of considerable palaeontological interest in view of their rich fossil biotas of terrestrial -like and other vertebrate groups as well as trace fossils (e.g. vertebrate burrows) that record the recovery of life on land following the end-Permian extinction. These fossil assemblages are assigned to the and Assemblage Zones and are best known from the Main Karoo Basin of .

Bedrock exposure of the Tarkastad along Section 3 of the R61 is generally poor and strongly biased towards road cuttings through more resistant-weathering sandstone packages as well as baked sediments adjacent to dolerite intrusions. Potentially fossiliferous successions are poorly represented. This applies especially to the more recessive-weathering Burgersdorp Formation, of which very few informative exposures were seen along the study route. Some of the road cuttings through the Katberg Formation (e.g. at km 38 and 41.6) and the Burgersdorp Formation (e.g. at km 70.6) are of sedimentological interest, and this would undoubtedly apply to the future widened cuttings as well. No vertebrate or plant body fossils were observed within the Tarkastad Subgroup rocks, which appear to be at most very sparsely fossiliferous along the study route. Few well-developed palaeosol (ancient soil) horizons marked by pedogenic calcrete nodules, with which vertebrate fossils are often associated, were seen here.

The only palaeontological remains recorded within the Tarkastad Subgroup bedrocks along Section 3 of the R61 were small invertebrate burrows – possibly the arthropod trace fossil 1 John E. Almond (2013) Natura Viva cc Katbergia - within mudrocks at one site (km 38) as well as possible large vertebrate burrows within overbank mudrocks at km 41.6. These Katburg Formation trace fossils are quite common and of fairly low heritage significance so no special conservation measures are proposed here. Good examples of large, shallowly inclined, sand- or mud-infilled vertebrate burrows excavated into the Katberg mudrocks are well seen in a road cutting close to Cradock (km 18.9), but this important, well-known locality lies some 5 km west of the present study area.

Late Caenozoic gravelly, silty and sandy alluvial deposits observed within river and stream banks en route are generally of low palaeontological sensitivity and no fossil or subfossil material was recorded therein.

All of the five proposed borrow pit sites (BP1-BP5) as well as the proposed new hard rock quarry (HRQ1) are underlain by Karoo dolerite intrusions and are of no palaeontological heritage significance. The dolerite in some cases is deeply weathered to yield resistant, rounded corestones embedded in friable sabunga. Adjacent sedimentary country rocks have been baked to quartzites and hornfels, seriously compromising their fossil heritage potential. No fossils were observed within these thermally metamorphosed country rocks.

It is concluded that the proposed upgrade of Section 3 of the R61 between Cradock and Tarkastad - including the proposed borrow pit and quarry excavations as well as modifications to stormwater structures - is of LOW palaeontological heritage significance. Pending the discovery of substantial new fossils during before or during development, no further specialist studies or mitigation in this respect are considered necessary for this road project.

Should substantial fossil remains be exposed during construction, however, such as vertebrate bones and teeth, plant-rich fossil lenses or dense fossil burrow assemblages, the Environmental Control Officer should safeguard these, preferably in situ, and alert ECPHRA (i.e. The Eastern Cape Provincial Heritage Resources Authority. Contact details: Mr Sello Mokhanya, 74 Alexander Road, King Williams Town 5600; [email protected]) as soon as possible so that appropriate action (e.g. recording, sampling or collection) can be taken by a professional palaeontologist. These recommendations should be incorporated into the Environmental Management Plan for the road project.

2. OUTLINE OF DEVELOPMENT

The South African National Roads Agency Limited (SANRAL) is proposing to rehabilitate Section 3 of the National Route R61 (km 24.2 to km 75) between Cradock and Tarkastad, Eastern Cape (Fig. 1). The project involves widening of the roadway by ± 5 m and also of all river or stream crossing structures along the route. These last comprise seven bridges (two to be raised by 1.5 m), six culverts and one pipe. All proposed roadworks will take place within the existing road reserve, with the exception of minor realigned sectors which will encroach onto farmland. Road material is to be sourced from five new or existing borrow pits and one hard rock quarry.

The present combined desktop and field-based palaeontological heritage assessment has been commissioned by Arcus GIBB (Pty) Ltd as part of the Basic Assessment of the proposed road development, in accordance with the requirements of the National Heritage Resources Act (Act 25 of 1999), and will also contribute to environmental management plans for the borrow pits and hard rock quarry developments (Contact details: Dr Norbert Klages, 2nd Floor, Greyville House, Cnr Greyville & Cape Rd, Greenacres, 6045; PO Box 63703, Greenacres 6057; Tel: (041) 392 7500; Fax: 041 363 9300; Email: [email protected]).

2 John E. Almond (2013) Natura Viva cc

Fig. 1. Google Earth© satellite image of the study area between Cradock and Tarkastad, Eastern Cape, showing the route of the R61 and the approximate locations of the five borrow pit sites (BP1 – BP5) and one hard rock quarry (HRQ1). Section 3 of the R61 (c. 50 km long) extends approximately from the BP1 marker eastwards to Tarkastad.

3 John E. Almond (2013) Natura Viva cc 2.1. National Heritage Resources Act

The extent of the proposed development (over 5000 m2 or linear development of over 300m) falls within the requirements for a Heritage Impact Assessment (HIA) as required by Section 38 (Heritage Resources Management) of the South African National Heritage Resources Act (Act No. 25 of 1999). The various categories of heritage resources recognised as part of the National Estate in Section 3 of the National Heritage Resources Act include, among others:

geological sites of scientific or cultural importance palaeontological sites palaeontological objects and material, meteorites and rare geological specimens

Minimum standards for the palaeontological component of heritage impact assessment reports have been developed by SAHRA (2013).

2.2. Approach used for this palaeontological study

The brief for the present palaeontological specialist study as defined by Arcus GIBB is as follows:

A phase one palaeontological impact assessment is to be undertaken and must address the following:

Assess each borrow pit and quarry site identified by the project engineers. On the R61 a total of 5 borrow pits and 1 hard rock quarry have been identified by the engineers. Areas where existing road cuttings are to be widened it must be assessed whether any heritage resources are likely to be affected or encroached upon. Where applicable, recommendations for the conservation of identified heritage resources are to be included.

This report provides a basic assessment of the observed or inferred palaeontological heritage within the Cradock – Tarkastad study area, with recommendations for any specialist palaeontological mitigation where this is considered necessary. The report is based on (1) a review of the relevant scientific literature, (2) geological maps, (3) previous palaeontological heritage assessments for other developments in the Karoo region (e.g. Almond 2010, 2011), (4) the author’s field experience with the formations concerned and their palaeontological heritage, and (5) a two-day field assessment on 19-20 December 2012 carried out by the author.

In preparing a palaeontological desktop study the potentially fossiliferous rock units (groups, formations etc) represented within the study area are determined from geological maps. The known fossil heritage within each rock unit is inventoried from the published scientific literature, previous palaeontological impact studies in the same region, and the author’s field experience (Consultation with professional colleagues as well as examination of institutional fossil collections may play a role here, or later during the compilation of the final report). This data is then used to assess the palaeontological sensitivity of each rock unit to development (Provisional tabulations of palaeontological sensitivity of all formations in the Western, Eastern and Northern Cape have already been compiled by J. Almond and colleagues; e.g. Almond et al. 2008). The likely impact of the proposed development on local fossil heritage is then determined on the basis of (1) the palaeontological sensitivity of the rock units concerned and (2) the nature of the development itself, most notably the extent of fresh bedrock excavation envisaged.

When rock units of moderate to high palaeontological sensitivity are present within the development footprint, a field assessment study by a professional palaeontologist is usually warranted. Most detrimental impacts on palaeontological heritage occur during the construction phase when fossils may be disturbed, destroyed or permanently sealed-in during excavations and subsequent construction activity. Where specialist palaeontological mitigation is recommended,

4 John E. Almond (2013) Natura Viva cc this may take place before construction starts or, most effectively, during the construction phase while fresh, potentially fossiliferous bedrock is still exposed for study. Mitigation usually involves the judicious sampling, collection and recording of fossils as well as of relevant contextual data concerning the surrounding sedimentary matrix. It should be emphasised that, provided appropriate mitigation is carried out, many developments involving bedrock excavation actually have a positive impact on our understanding of local palaeontological heritage. Constructive collaboration between palaeontologists and developers should therefore be the expected norm.

The focus of the field-based assessment work is not simply to survey the development footprint or even the development area as a whole (e.g. farms or other parcels of land concerned in the development). Rather, the palaeontologist seeks to assess or predict the diversity, density and distribution of fossils within and beneath the study area, as well as their heritage or scientific interest. This is primarily achieved through a careful field examination of one or more representative exposures of all the units present (N.B. Metamorphic and igneous rocks rarely contain fossils). The best rock exposures are generally those that are easily accessible, extensive, and fresh (i.e. unweathered) and include a large fraction of the stratigraphic unit concerned (e.g. formation). These exposures may be natural or artificial and include, for example, rocky outcrops in stream or river banks, cliffs, quarries, dams, dongas, open building excavations or road and railway cuttings. Uncemented superficial deposits, such as alluvium, scree or wind-blown sands, may occasionally contain fossils and should also be included in the scoping study where they are well-represented in the study area. It is normal practice for impact palaeontologists to collect representative, well-localized (e.g. GPS and stratigraphic data) samples of fossil material during field assessment studies. However, fossil collection should be supported by a permit from the relevant heritage heritage authority and all fossil material collected must be properly curated within an approved repository (usually a museum or university collection).

Before fieldwork commenced, a preliminary screening of satellite images and 1: 50 000 maps of the R61 study area was conducted to identify any sites of potentially good bedrock exposure to be examined in the field. These sites might include, for example, natural exposures (e.g. stream beds, rocky slopes, stream gullies) as well as artificial exposures such as quarries, dams and cuttings along farm tracks.

Note that while fossil localities recorded during fieldwork within the study area itself are obviously highly relevant, most fossil heritage here is embedded within rocks beneath the land surface or obscured by surface deposits (soil, alluvium etc) and by vegetation cover. In many cases where levels of fresh (i.e. unweathered) bedrock exposure are low, the hidden fossil resources have to be inferred from palaeontological observations made from better exposures of the same formations elsewhere in the region but outside the immediate study area. Therefore a palaeontologist might reasonably spend far more time examining road cuts and borrow pits close to, but outside, the study area than within the study area itself. Field data from localities even further afield (e.g. an adjacent province) may also be adduced to build up a realistic picture of the likely fossil heritage within the study area.

On the basis of the desktop and field assessment studies, the likely impact of the proposed development on local fossil heritage and any need for specialist mitigation are then determined. Adverse palaeontological impacts normally occur during the construction rather than the operational or decommissioning phase. Mitigation by a professional palaeontologist – normally involving the recording and sampling of fossil material and associated geological information (e.g. sedimentological data) – is usually most effective during the construction phase when fresh fossiliferous bedrock has been exposed by excavations, although pre-construction recording of surface-exposed material may sometimes be more appropriate. To carry out mitigation, the palaeontologist involved will need to apply for a palaeontological collection permit from the relevant heritage management authority (i.e. The Eastern Cape Provincial Heritage Resources Authority or ECPHRA. Contact details: Mr Sello Mokhanya, 74 Alexander Road, King Williams Town 5600; [email protected]). It should be emphasized that, providing appropriate mitigation is carried out, the majority of developments involving bedrock excavation can make a positive contribution to our understanding of local palaeontological heritage. 5 John E. Almond (2013) Natura Viva cc 2.3. Assumptions & limitations

The accuracy and reliability of palaeontological specialist studies as components of heritage impact assessments are generally limited by the following constraints:

1. Inadequate database for fossil heritage for much of the RSA, given the large size of the country and the small number of professional palaeontologists carrying out fieldwork here. Most development study areas have never been surveyed by a palaeontologist.

2. Variable accuracy of geological maps which underpin these desktop studies. For large areas of terrain these maps are largely based on aerial photographs alone, without ground-truthing. The maps generally depict only significant (“mappable”) bedrock units as well as major areas of superficial “drift” deposits (alluvium, colluvium) but for most regions give little or no idea of the level of bedrock outcrop, depth of superficial cover (soil etc), degree of bedrock weathering or levels of small-scale tectonic deformation, such as cleavage. All of these factors may have a major influence on the impact significance of a given development on fossil heritage and can only be reliably assessed in the field.

3. Inadequate sheet explanations for geological maps, with little or no attention paid to palaeontological issues in many cases, including poor locality information.

4. The extensive relevant palaeontological “grey literature” - in the form of unpublished university theses, impact studies and other reports (e.g. of commercial mining companies) - that is not readily available for desktop studies.

5. Absence of a comprehensive computerized database of fossil collections in major RSA institutions which can be consulted for impact studies. A Karoo fossil vertebrate database is now accessible for impact study work.

In the case of palaeontological desktop studies without supporting Phase 1 field assessments these limitations may variously lead to either:

(a) underestimation of the palaeontological significance of a given study area due to ignorance of significant recorded or unrecorded fossils preserved there, or

(b) overestimation of the palaeontological sensitivity of a study area, for example when originally rich fossil assemblages inferred from geological maps have in fact been destroyed by tectonism or weathering, or are buried beneath a thick mantle of unfossiliferous “drift” (soil, alluvium etc).

Since most areas of the RSA have not been studied palaeontologically, a palaeontological desktop study usually entails inferring the presence of buried fossil heritage within the study area from relevant fossil data collected from similar or the same rock units elsewhere, sometimes at localities far away. Where substantial exposures of bedrocks or potentially fossiliferous superficial sediments are present in the study area, the reliability of a palaeontological impact assessment may be significantly enhanced through field assessment by a professional palaeontologist.

The main limitation during the present field-based basic assessment of palaeontological heritage along the R61 between Cradock and Tarkastad was the low level of fresh bedrock exposure, especially within the outcrop area of the Burgersdorp Formation. Cuttings tend to be developed where the road intersects resistant-weathering dolerite intrusions and thick sandstone packages (forming topographic highs) while the potentially fossilferous mudrock units are seriously under- represented since they weather more easily and therefore generate little topographic relief.

6 John E. Almond (2013) Natura Viva cc

BP5 BP4

HRQ1 BP3

BP2

BP1

N

10 km

Fig. 2. Composite mosaic of extracts from adjoining 1: 250 000 geology sheets 3124 Middelburg, 3126 Queenstown, 3226 King Williams Town and 3224 Graaff-Reinet, in clockwise order from the top left (Council for Geoscience, Pretoria). The western sector of the R61, towards Cradock, is underlain by the Katberg Formation (TRk) while the eastern sector is underlain by the Burgersdorp Formation (TRb). The contact between these units lies on the Middelburg sheet (blue arrow) and is obscured by alluvium. The upper upper Beaufort Group sediments are extensively intruded and baked by dolerite sills and dykes (Jd). The red / orange triangle symbols indicate the approximate location of the five borrow pit sites (BP1-BP5) and one hard rock quarry site (HRQ1) along the N61 between Cradock and Tarkastad, all excavated into dolerite. 7 John E. Almond (2013) Natura Viva cc 2. GEOLOGICAL CONTEXT

The study area between Cradock and Tarkastad is a semi-arid, undulating hilly region between the Bamboesberg to the north and the Winterberg to the south within the eastern portion of the Great Karoo sensu lato. This lies within the Eastern and South-Eastern Middleveld physiographic region of South Africa (Visser 1989). The following outline of the topographic setting for Section 3 of the R61 (between BP1 and Tarkastad on Figs. 1 and 2) has kindly been supplied by Arcus GIBB (Pty) Ltd, Port Elizabeth:

From its start at Dwingfontein (altitude 1080 m) Section 3 of the R61 traverses a gently undulating flat until reaching the Vlekpoortrivier at Klipkraal where it passes through a gap between two mountains of 1200 m in height. Thereafter, in long straight sections, the road climbs slightly, passing the Maermansberg (1674 m) on the right side, until it reaches the crossing of the Elandsrivier at the 401 turnoff to (height 1250 m). On its way to Tarkastad the R61 stays at the valley bottom between the Elandskop peak (1749 m) and the spectacular buttress of the Middelkraal mountains (2031 m). The road then climbs steadily until it reaches the town limit of Tarkastad at a height of 1320 m.

The geology of the R61 study area is outlined in the 1: 250 000 geology sheets 3124 Middelburg, 3126 Queenstown, 3226 King William’s Town and 3224 Graaff-Reinet (Fig. 2). Geological sheet explanations for these sheets are provided by Cole et al. (2004), Johnson (1984) and Hill (1993) (A very short explanation to the King William’s Town map is printed on the map sheet itself). Section 3 of the R61 traverses the outcrop area of continental sedimentary rocks of the Upper Beaufort Group (Tarkastad Subgroup, ) of Early to age (Johnson et al. 2006). These are intruded by numerous dolerite intrusions of the Karoo Dolerite Suite of Early Jurassic age. Towards Cradock (Graaff-Reinet and Middelburg 1: 250 000 sheet areas) the sedimentary bedrocks belong to the sandstone-dominated Katberg Formation that was deposited in arid braided fluvial settings following the catastrophic end-Permian mass extinction event. Further east towards Tarkastad (Queenstown and King William’s Town 1: 250 000 sheet areas) the sedimentary bedrocks are assigned to the slightly younger Burgersdorp Formation comprising recessive-weathering reddish mudrocks and braided river channel sandstones. The contact between the Katberg and Burgersdorp Formations along the R61 lies on the Middelburg sheet (blue arrow near Klipkraal in Fig. 2) where it is obscured by alluvium. Late Caenozoic superficial deposits such as colluvium (scree etc), alluvium, various gravels and soils mantle considerable portions of the Tarkastad Subgroup outcrop area, especially where the successions are dominated by more easily-weathered and eroded mudrocks. In general levels of sedimentary bedrock exposure in the Eastern Cape study area are poor, due to this superficial mantle as well as vegetation cover. Informative exposures are mainly confined to occasional road cuttings, incised river banks and steeper hillslopes as well as farm dams.

GPS data for all localities mentioned in the text are provided in the Appendix.

2.1. Katberg Formation

Useful geological descriptions of the Katberg Formation, the basal subunit of the Tarkastad Subgroup in the study area, are given by Johnson (1976), Hancox (2000), Johnson et al. (2006), Smith et al. (2002) and for the Middelburg and Graaff-Reinet sheet areas in particular by Cole et al. (2004) and Hill (1993) respectively. The more detailed sedimentological accounts of the Katberg rocks by Stavrakis (1980), Hiller and Stavrakis (1980, 1984), Haycock et al. (1994), Groenewald (1996) and Neveling (1998) are also relevant to the present study area.

The Katberg Formation forms the regionally extensive, sandstone-rich lower portion of the Tarkastad Subgroup (Upper Beaufort Group) that can be traced throughout large areas of the Main Karoo Basin. In the Middelburg sheet area it reaches a maximum thickness of some 400 m, but close to Noupoort thicknesses of 240-260 m are more usual. The predominant sediments are (a) 8 John E. Almond (2013) Natura Viva cc prominent-weathering, pale buff to greyish, tabular or ribbon-shaped sandstones up to 60 m thick that are interbedded with (b) recessive-weathering, reddish or occasionally green-grey mudrocks. Up to four discrete sandstone packages can be identified within the succession. Katberg channel sandstones are typically rich in and lithic grains (i.e. lithofeldspathic). They build laterally extensive, multistorey units with an erosional, often gullied base that is usually marked by intraformational conglomerates or parabreccias up to one meter thick consisting of mudrock pebbles, reworked calcrete nodules and occasional rolled fragments of bone. Internally the moderately well-sorted sandstones are variously massive, horizontally-laminated or cross-bedded and heavy mineral laminae occur frequently. Sphaeroidal carbonate concretions up to 10 cm across are common. The predominantly reddish Katberg mudrocks are typically massive with horizons of pedocrete nodules (calcretes), and mudcracks. Mudrock exposure within the study area is very limited due to extensive mantling of these recessive-weathering rocks by superficial sediments.

Sandstone deposition was mainly due to intermittently flooding, low-sinuosity braided river systems flowing northwards from the rising mountains in the south into the subsiding Main Karoo Basin (Fig. 3). Mudrocks were largely laid down by suspension settling within overbank areas following episodic inundation events, while other fine-grained sediments are associated with lakes and temporary playas in lower-lying areas on the arid floodplain, especially in the northern Katberg outcrop area and its lateral correlatives in the Burgersdorp Formation. Palaeoclimates inferred for the Period in the Main Karoo Basin were arid with highly seasonal rainfall and extensive periods of drought. This is suggested by the abundant oxidised (“rusty red”) mudrocks, desiccation cracks, and palaeosols associated with well-developed calcretes. Arid settings are also supported by taphonomic and behavioural evidence such as pervasive carbonate encrustation of fossil bones, mummification of postcrania, bone-bed death assemblages associated with water holes and the frequency of burrowing habits among tetrapods, including large like Lystrosaurus (Groenewald 1991, Smith & Botha 2005, Viglietti 2010).

The Katberg Formation is well exposed in a number of road cuttings along the R61 between Cradock and Klipkraal. Spectacular long, deep cuttings at around km 18.9 (Locs. 067, 068) – to the west of the present study area - have already been described in the literature and show classic sedimentological features of the formation as well as important trace fossil assemblages (Bordy et al. 2010). Within the study area especially good exposures of Katberg channel sandstones are seen east of Vlekpoort (Loc. 080, Fig. 4). Well-sorted, buff, feldspathic sandstones build multi- storey packages (6 m or more thick) and various show massive, horizontally-laminated to cross- laminated fabrics and tabular to lenticular geometries (Fig. 5). Horizontally-laminated sandstone facies may show primary current lineation and well-developed, finely-spaced heavy mineral lamination (e.g. Loc. 077). The bases of the sandstones are erosive, locally gullied, and associated with thin to thick (30 cm) parabreccias dominated by reworked angular to subrounded mudrock intraclasts and calcrete glaebules in a sandy matrix (Figs. 10 & 11). Sandstones may contain dispersed ovoid, resistant-weathering diagenetic concretions (possibly secondarily silicified) (Fig. 6) and “floating” angular mudrock intraclasts (Fig. 10). Occasional meter-scale beds of fine- grained, tabular cross-bedded gravels composed largely of mudflakes point towards channel infill during episodes of catastrophic floodplain erosion, perhaps associated with protracted drought (Loc. 080, Fig. 7).

9 John E. Almond (2013) Natura Viva cc

Fig. 3. Reconstruction of the south-eastern part of the Main Karoo Basin in Early to Middle Triassic times showing the deposition of the Tarkastad Subgroup. The sandstone- dominated succession of the Katberg Formation was deposited in near the mountainous source area in the south. The overlying to interfingering mudrock-dominated Burgersdorp Formation was deposited on the distal floodplain where numerous playa lakes are also found (From Hiller & Stavrakis 1984).

Thick purple-brown to (more rarely) grey-green mudrock packages contain well-developed, downward-tapering sand-infilled desiccation cracks up to 30 cm deep, locally associated with large vertebrate burrows with an ovoid cross-section (Figs. 9 & 39). The mudrocks are massive to thin- bedded and may contain small-scale lenticular, sandstone-infilled floodplain channels. Horizons of chaotically intermixed mudrock and sandstone perhaps reflect channel bank collapse during major flood events. Large flame structures, boudinaged sandstone beds and rounded pseudonodules result from loading and dewatering processes. Large, gently-inclined vertebrate burrows seen here are briefly described below in Section 3.1 (Figs. 8 & 38). In general, well-developed palaeosol horizons marked by abundant pedogenic calcrete nodules are apparently not common in the study area (e.g. Locs. 172, 073, 076) and are often secondarily modified - ferruginised and / or silicified - by dolerite intrusion (Fig. 37). Colour-banded road cut exposures comprising thin-bedded mudrocks and thin, tabular sandstones probably represent distal floodplain or even playa lake facies (Fig. 12, Loc. 076). The mudrocks here contain numerous small calcrete nodules as well as small-scale burrows of probable invertebrate origin (Figs. 13 & 49).

10 John E. Almond (2013) Natura Viva cc

Fig. 4. Well-exposed vertical section through a Katberg sandstone package showing erosive-based, tabular to lenticular channel sandstones and subordinate interbeds of greenish-grey mudrock (Loc. 080).

Fig. 5. Multi-storey package of tabular to lenticular buff channel sandstones of the Katberg Formation showing frequent interbeds of mudclast-rich parabreccias (Loc. 077). Cross- bedding within channel sandstones in this area (not shown here) indicates SW-directed palaeocurrents.

11 John E. Almond (2013) Natura Viva cc

Fig. 6. Sphaeroidal diagenetic nodules, secondarily ferruginised and probably originally siliceous, within the Katberg channel sandstones shown in the previous figure. Nodules seen here are up to 15 cm across and are unfossiliferous.

Fig. 7. Thick, well-consolidated bed of cross-laminated fine-grained mudflake gravels (secondarily silicified) suggesting episodic catastrophic erosion of the Katberg floodplain (Loc. 080) (Hammer = 27 cm).

12 John E. Almond (2013) Natura Viva cc

Fig. 8. Interbedded thick buff, tabular to lenticular channel sandstones and purple-brown overbank mudrocks of the Katberg Formation (Loc. 067, R61 west of study area).

Fig. 9. Thin crevasse-splay sandstones overlying wedge-shaped sandstone-infilled desiccation cracks within overbank mudrocks. The mudrocks are capped by a channel sandstone unit containing dispersed large reworked mudrock intraclasts, Katberg Formation (Loc. 067).

13 John E. Almond (2013) Natura Viva cc

Fig. 10. Close-up of basal channel parabreccia with poorly-sorted, angular to subrounded intraclasts of purplish-brown and greenish mudrock, Katberg Formation (Loc. 068) (Hammer = 27 cm). Reworked disarticulated fossil bones are occasionally found in such channel parabreccias elsewhere in the Katberg Formation outcrop area.

Fig. 11. Massive to horizontally-laminated channel sandstones of the Katberg Formation showing narrow basal gullies incised into the underlying purple-brown, massive, hackly- weathering overbank mudrocks (Loc. 073). The mudrocks contain dispersed pedogenic calcrete nodules (not clearly visible in photo) and greenish reduction spots.

14 John E. Almond (2013) Natura Viva cc

Fig. 12. Probable distal floodplain or playa lake deposits of the Katberg Formation comprising thin-bedded purple-brown mudrocks and paler crevasse splay sandstones (Loc. 076, between Fairview and Vlekpoortrivier).

Fig. 13. Detail of the exposure seen in the previous figure showing abundant small, spheroidal, ferruginised calcrete nodules within the thin-bedded mudrock intervals (Hammer = 27 cm) (Loc. 076).

15 John E. Almond (2013) Natura Viva cc 2.2. Burgersdorp Formation

The Burgersdorp Formation is the youngest subunit of the Permo-Triassic Beaufort Group (Karoo Supergroup) and is paraconformably overlain by the Molteno and Elliot Formations of the . It is a mudrock-rich succession of Early to Middle Triassic age with a total thickness of some 900-1000m in its southern outcrop area near Queenstown (Johnson et al. 2006). Brief geological descriptions of the Burgersdorp Formation are given by Karpeta and Johnson (1979), Dingle et al. (1983), Johnson (1976, 1984), Hiller & Stavrakis (1984), Johnson & Hiller (1990), Kitching (1995) and Hancox (2000; see also extensive references therein). There are also excellent sedimentological accounts of these rocks in the Karoo excursion guides edited by Smith et al. (1998, 2002).

The Burgersdorp rocks were laid down within the Main Karoo Basin by northwestwards-flowing meandering rivers during a warm, arid to semi-arid climatic interval (Fig. 3). They comprise isolated, lenticular, feldspathic channel sandstones, abundant crevasse splay sandstones, and typically greyish-red to dusky red overbank mudrocks, forming upwards-fining cycles of a few meters to tens of meters in thickness. Intraformational mudflake breccio-conglomerates (often described as matrix-rich parabreccias) are common at the base of the sandstone units. The mudrocks are generally massive (unbedded) but occasionally display sand-infilled mudcracks and clastic dykes. Well-laminated reddish mudrocks with pedocrete (cemented soil) horizons are interpreted as playa lake deposits. Lacustrine palaeoenvironments predominated in the northern part of the Karoo Basin at this time and these lake deposits have recently received considerable palaeontological attention (e.g. Free State; Welman et al. 1995, Hancox et al. 2010 and refs therein).

In general the Burgersdorp Formation along Section 3 of Route 61 is very poorly exposed indeed, largely due the predominance of readily-weathered and -eroded mudrock facies. Thick successions of reddish-brown mudrocks and prominent-weathering sandstone packages of the Burgersdorp Formation are seen on steeper mountain slopes in the Tarkastad region, below the dolerite capping – for example in the Middelkraal Mountains (Fig. 14). Purplish-grey speckled channel sandstones at Loc. 083 contain ferruginous concretions and large subrounded intraclasts of reddish-brown mudrock (Fig. 16). In a roadside gulley at Loc. 086a a succession of greenish- grey Burgersdorp and thin- to medium-bedded buff sandstones with rib-and-furrow ripple lamination contains a 50-60 cm – thick parabreccia bed containing fine mudrock intraclasts and small calcrete glaebules (Fig. 17). An extensive series of road cuttings along the R61 just west of Tarkastad (Loc. 088) exposes vertical sections through buff tabular sandstone packages with minor mudrock interbeds. The sandstones often display well-developed horizontal lamination (Fig. 15) and silicified basal mudclast parabreccias where these lie within the thermal aureole of a dolerite intrusion (Fig. 16). Low river banks near the bridge at Loc. 087 expose colour-banded, purple-brown, hackly-weathering, thin-bedded mudrocks of the Burgersdorp Formation capped by a thin, erosive-based channel sandstone (Fig. 18). The mudrocks contain small calcrete nodules.

16 John E. Almond (2013) Natura Viva cc

Fig. 14. Reddish Burgersdorp red beds exposed on the steep lower slopes of the Middelkraal mountains north of Tarkastad, viewed towards the north. The mountain is capped by a thick dolerite sill showing vertical columnar jointing.

Fig. 15. Tabular, horizontally-laminated channel sandstones of the Burgersdorp Formation with hackly-weathering, khaki mudrock interbeds (Loc. 088) (Hammer = 27 cm).

17 John E. Almond (2013) Natura Viva cc

Fig. 16. Horizons rich in angular, secondarily silicified mudrock intraclasts within Burgersdorp sandstones (Loc. 088) (Hammer = 27 cm).

Fig. 17. Medium-bedded, massive horizon on fine mudrock and calcrete gravels, erosion gulley within the Burgersdorp Formation (Loc. 086a) (Hammer = 27 cm).

18 John E. Almond (2013) Natura Viva cc

Fig. 18. Rare riverbank exposure of thin-bedded reddish-brown mudrocks of the Burgersdorp Formation capped by a thin, lenticular channel sandstone and poorly-sorted Late Caenozoic alluvial gravels (Loc. 087) (Hammer = 27 cm).

2.3. Karoo Dolerite Suite

The Triassic Tarkastad Subgroup sediments within the study area are extensively intruded and thermally metamorphosed (baked) by subhorizontal sills and steeply inclined dykes of the Karoo Dolerite Suite (Jd). Good examples in the study region include the major sills intruding Katberg sediments at the Vlekpoortrivier (Klipkraal area, Fig. 19) as well as the prominent sill capping the Middelkraal mountain range north of Tarkastad (Fig. 14). These Early Jurassic (c. 183 Ma) basic intrusions were emplaced during crustal doming and stretching that preceded the break-up of Gondwana (Duncan and Marsh 2006). The hot dolerite magma baked adjacent Beaufort Group mudrocks and sandstones to form dark grey, splintery hornfels and quartzites respectively.

Karoo dolerite intrusions, including both steeply-inclined to subvertical dykes as well as more gently inclined, subconformable to transgressive sills, are well-represented in road cuttings and river exposures along the R61. The associated Tarkastad sandstones are baked to well-jointed, pale quartzites and splintery hornfels, also increasing their resistance to erosion. Good cliff exposures of a thick dolerite sill are seen at Vlekpoort (Loc. 079), showing the blocky jointing of the fresher dolerite beneath and corestones towards the surface (Fig. 19). The underlying Katberg sandstones (now altered to metaquartzite) within the thermal aureole of the sill at Vlekpoort also form steep cliffs (Fig. 20). Contacts between the Tarkastad Subgroup country rocks and the intrusive dolerites are usually sharp (Figs. 21 & 22), and in some cases may have been modified by faulting (possibly the case at Loc. 088 near Tarkastad, for example). At the last locality a thin late-stage felsic dyke cross-cutting the main dolerite intrusion has been displaced by a small-scale, subvertical normal fault (Fig. 24).

19 John E. Almond (2013) Natura Viva cc

Fig. 19. Masonry-like blocky joining within a thick dolerite sill, Vlekpoort (Loc. 079).

Fig. 20. Thick package of themally metamorphosed, resistant-weathering Katberg quartzites (pale grey) within the thermal aureole of an overlying major dolerite sill (rusty brown) at Vlekpoort (Loc. 079).

20 John E. Almond (2013) Natura Viva cc

Fig. 21. Sharp, oblique contact between dark grey-green Karoo dolerite (left) and interbedded pale grey Katberg quartzites with interbeds of mudclast parabreccia (right), Loc. 070 near Dwingfontein quarry site (BP1) (Hammer = 27 cm).

Fig. 22. Narrow (< 1 m), steeply inclined dyke of dolerite intruding purple-brown Katberg mudrocks at Loc. 072. The wall rocks adjacent to the dyke are baked and prominent- weathering. Pedogenic calcretes within the mudrocks (e.g. to left of hammer) are secondarily ferruginised and silicified (Hammer = 27 cm).

21 John E. Almond (2013) Natura Viva cc

Fig. 23. Low roadside cliff of massive, well-jointed dark Katberg hornfels (baked mudrock) capped by a thin paler quartzite bed (baked sandstone) and overlain by a thick dolerite sill (Loc. 074) (Hammer = 27 cm).

Fig. 24. Major dolerite intrusion within the Burgersdorp Formation near Tarkastad traversed by a later thin pale (felsic) dyke that has in turn been displaced by a minor normal fault (Loc. 088) (Hammer = 27 cm).

22 John E. Almond (2013) Natura Viva cc

Baked quartzites are usually brittle, well-jointed, and pale grey, sometimes spotted or mottled (e.g. at Loc. 074). Interbedded parabreccias are frequently vuggy (vesicular) and the included mudrock and calcrete intraclasts are often secondarily ferruginised. Hornfelsed mudrocks are often spotted, brittle and splintery and also well-jointed (Fig. 23). Pedogenic calcrete nodules (some of which may originally have been fossiliferous) are usually secondarily ferruginised and may be silicified (See Figs. 22 & 37).

All the proposed borrow pit sites (Figs 1 and 2, BP1 – BP5) as well as the hard rock quarry (HRQ1) are underlain by Karoo dolerite, in some cases with adjacent thermally metamorphosed country rocks.

BP1 (Dwingfontein) is an existing quarry excavated into deeply-weathered dolerite on the eastern flank of a NNW-SSE trending, ENE-dipping dolerite ridge. Subrounded to subangular corestones of fresher dolerite are embedded within friable, deeply-rotten dolerite or sabunga, with a rusty discoloration near-surface (Fig. 25). The ridge is mantled by rubbly dolerite corestone colluvium; many of the boulders show a shiny black patina of desert varnish (often a favoured palette for rock engravings in the Karoo region). A thin veneer of calcrete has developed along joint fractures, especially at the contact between dolerite bedrock and weathered regolith. Fresh dolerite road cuttings show angular, blocky to undulose sheet-jointing as well as typical dolerite onion-skin weathering near-surface.

BP2 (Fairview) borrow pit site overlying a NW-SE trending dolerite intrusion is mantled with bouldery and gravelly dolerite colluvium and soils, with evidence of previous mining further from the road. Tarkastad (Katberg) sedimentary bedrocks are not seen on site but a well-exposed in road cuttings just to the north (Loc. 076, Figs. 12 & 13).

BP3 (Klipkraal) is a new borrow pit site on gently-sloping ground underlain by weathered dolerite with in situ rounded corestones passing up into foliated sabunga and orange-brown doleritic gravelly soils and downwasted corestones, as exposed in adjacent road cuttings (Fig. 26).

BP4 (Burnley Park) is an extension of an existing pit into weathered dolerite (Fig. 27) that shows evidence of late stage intrusive veining and is mantled with disturbed alluvial pediment gravels, including well-rounded sandstone and hornfels clasts.

BP5 (Prinsfontein) is also an extension of a large existing pit showing good vertical sections through deeply-weathered sabunga overlain by orange-brown calcretised soils and gravels (Fig. 33).

The proposed hard rock quarry site (HRQ1) at Fairview overlies a major NW-SE trending dolerite intrusion near the Vlekpoortrivier. Good exposures of fresh, coarse-grained, speckled dolerite with sheet-like jointing are well-exposed in stream beds in the area (Fig. 28). The adjacent Katberg Formation country rocks have been extensively baked and silicified to pale buff quartzites and darker grey hornfels here.

23 John E. Almond (2013) Natura Viva cc

Fig. 25. Weathered dolerite at BP1 (Dwingfontein) with ferruginous discoloration near- surface (Hammer = 27 cm).

Fig. 26. Road cutting through weathered, foliated dolerite sabunga and corestones adjacent to BP3 (Klipkraal) (Hammer = 27 km).

24 John E. Almond (2013) Natura Viva cc

Fig. 27. Thick section through weathered dolerite (sabunga) in BP4 borrow pit (Burnley Park), mantled by rubbly down-wasted dolerite and pediment gravels (Hammer = 27 cm).

Fig. 28. Sheet-like jointing of coarse-grained dolerite exposed in stream banks near the proposed hard rock quarry site at Fairview (HRQ1).

25 John E. Almond (2013) Natura Viva cc 2.4. Late Caenozoic superficial deposits

Various types of superficial deposits (“drift”) of Late Caenozoic (largely Quaternary to Recent) age occur widely throughout the Karoo region, including in the study area between Cradock and Tarkastad. They include minor pedocretes (e.g. calcretes or soil limestones), colluvial slope deposits (sandstone and dolerite scree, downwasted gravels etc), sheet wash, river channel alluvium and terrace gravels, as well as spring and pan sediments (Cole et al., 2004, Partridge et al. 2006). Only the larger tracts of Quaternary to Recent alluvium overlying the Tarkastad Subgroup are shown on the 1: 250 000 scale geological maps (pale yellow with “flying ” symbol in Fig. 2). In practice, a high proportion of the Karoo sedimentary bedrocks are in fact mantled by superficial deposits in the broader study region, especially in lower-lying regions underlain by mudrock-rich successions.

Several meters of silty and andy alluvium as well as fine stream gravels and reworked boulder dolerite corestones are seen, for example, in the dry stream bed adjacent to the road bridge at Loc. 071 (km 26), while silty alluvium predominates in the shallow stream bed.at km 36.9 (Fig. 29). Vertical sections through thick, brownish, crudely- to well-bedded gravelly, sandy alluvium are well seen overlying dolerite and Katberg hornfels in the Vlekpoortrivier at km 46 (Fig. 30, Loc. 082) and overlying Burgersdorp sandstones at km 61.4 (Loc. 084b, Fig. 31). Poorly-sorted alluvial gravels, predominantly of subangular sandstone clasts, mantle Burgersdorp Formation sediments at km 67.2 (Loc. 087, Fig. 18).

Dolerite subgroup is usually overlain by downwasted dolerite corestones (Fig. 32) and finer colluvial gravels and orange-brown sandy soils, locally consolidated with calcrete cement (Fig. 33). Downwasted surface gravels throughout the study region consist predominantly of tougher such as dolerite, Tarkastad sandstone, pale quartzite and dark hornfels. Clasts of the last type (“lydianite”) are frequently flaked (Fig. 34).

Fig. 29. Thick silty alluvium completely mantles Tarkastad bedrock in the river bed at km 36.9.

26 John E. Almond (2013) Natura Viva cc

Fig. 30. Gravelly alluvium overlying Katberg hornfels and dolerite at Vlekpoortrivier bridge, km 46 (Loc. 082).

Fig. 31. Well-bedded, fine gravelly and sandy alluvium overlying Burgersdorp sandstones in the river bed at km 61.4 (Loc. 084b).

27 John E. Almond (2013) Natura Viva cc

Fig. 32. Downwasted dolerite corestones and gravelly soils overlying dolerite bedrock at the Klipkraal (BP3) borrow pit site (Loc. 081).

Fig. 33. Extensively calcretised, orange-brown sandy soils with sparse gravels (mainly dark hornfels) mantling deeply-weathered dolerite at the Prinsfontein (BP5) borrow pit site (Hammer = 27 cm).

28 John E. Almond (2013) Natura Viva cc

Fig. 34. Sparse downwasted surface gravels overlying dolerite at Klipkraal (Loc. 081) showing predominance of dark hornfels clasts, some of which are artificially flaked (Hammer = 27 cm).

29 John E. Almond (2013) Natura Viva cc 3. PALAEONTOLOGICAL HERITAGE

The fossil heritage that has been previously recorded from the four main rocks units that are represented in the R61 study area between Cradock and Tarkastad is briefly outlined here. GPS data for all localities mentioned in the text are provided in the Appendix.

3.1. Fossils within the Katberg Formation

The Katberg Formation is known to host a low-diversity but palaeontologically important terrestrial fossil biota of Early Triassic (Scythian / Induan - Early Olenekian) age, i.e. around 250 million years old (Groenewald & Kitching 1995, Rubidge 2005). The biota is dominated by a small range of (“mammal-like reptiles”), amphibians and other tetrapods, with rare vascular plants and trace fossils, and has been assigned to the Lystrosaurus Assemblage Zone (LAZ). This impoverished fossil assemblage characterizes Early Triassic successions of the upper part of the Palingkloof Member (Adelaide Subgroup) as well as the Katberg Formation and - according to some earlier authors – the lowermost Burgersdorp Formations of the Tarkstad Subgroup. Recent research has emphasized the rapidity of faunal turnover during the transition between the sand- dominated Katberg Formation (Lystrosaurus Assemblage Zone) and the overlying mudrock- dominated Burgersdorp Formation (Cynognathus Assemblage Zone) (Neveling et al. 2005). In the proximal (southern) part of the basin the abrupt faunal turnover occurs within the uppermost sandstones of the Katberg Formation and the lowermost sandstones of the Burgersdorp Formation (ibid., p.83 and Neveling 2004). This work shows that the Cynognathus Assemblage Zone correlates with the entire Burgersdorp Formation; previous authors had proposed that the lowermost Burgersdorp beds belonged to the Lystrosaurus Assemblage Zone (e.g. Keyser & Smith 1977-78, Johnson & Hiller 1990, Kitching 1995). It should also be noted that the Lystrosaurus has now been recorded from the uppermost beds of the Latest Permian Dicynodon Assemblage Zone but only becomes super-abundant in Early Triassic times (e.g. Smith & Botha 2005, Botha & Smith 2007 and refs. therein).

Useful illustrated accounts of LAZ fossils are given by Kitching (1977), Keyser and Smith (1977- 1978), Groenewald and Kitching (1995), MacRae (1999), Hancox (2000), Smith et al. (2002), Cole et al. (2004), Rubidge (2005 plus refs therein) and Damiani et al. (2003a), among others. These fossil biotas are of special palaeontological significance in that they document the recovery phase of terrestrial ecosystems following the catastrophic end-Permian Mass Extinction of 251.4 million years ago (e.g. Smith & Botha 2005, Botha & Smith 2007 and refs. therein). They also provide interesting insights into the adaptations and taphonomy of terrestrial and plants during a particularly stressful, arid phase of Earth history in the Early Triassic.

Key tetrapods in the Lystrosaurus Assemblage Zone biota are various species of the medium- sized, shovel-snouted dicynodont Lystrosaurus (by far the commonest fossil form in this biozone. contributing up to 95% of fossils found), the small captorhinid parareptile , the crocodile-like early , and a wide range of small to large armour-plated “labyrinthodont” amphibians such as (Figs. 35 & 36). Botha and Smith (2007) have charted the ranges of several discrete Lystrosaurus species either side of the Permo-Triassic boundary. Also present in the LAZ are several genera of small-bodied true reptiles (e.g. owenettids), therocephalians, and early (e.g. , ). burrows are attributable to various aquatic and land-living invertebrates, including arthropods (e.g. Scoyenia scratch burrows), as well as several subgroups of fossorial tetrapods such as cynodonts, procolophonids and even Lystrosaurus itself (e.g. Groenewald 1991, Damiani et al. 2003b, Abdala et al. 2006, Modesto & Brink 2010, Bordy et al. 2009, 2011). Vascular plant fossils are generally rare and include petrified wood (“Dadoxylon”) as well as leaves of glossopterid progymnosperms and arthrophyte ferns (Schizoneura, Phyllotheca). An important, albeit poorly-preserved, basal Katberg palaeoflora has recently been documented from the Noupoort area (Carlton Heights) by Gastaldo et al. (2005). Plant taxa here include sphenopsid axes, dispersed fern pinnules and possible peltasperm (seed fern) reproductive structures. Pebbles of reworked silicified wood of 30 John E. Almond (2013) Natura Viva cc possible post- age occur within the Katberg sandstones in the proximal outcrop area near East London (Hiller & Stavrakis 1980, Almond unpublished obs.). Between typical fossil assemblages of the Lystrosaurus and Cynognathus Assemblage Zones lies a possible Procolophon Acme Zone characterized by abundant material of procolophonids and of the amphibian Kestrosaurus but lacking both Lystrosaurus and Cynognathus (Hancox 2000 and refs. therein).

Most vertebrate fossils are found in the mudrock facies rather than channel sandstones. Articulated skeletons enclosed by calcareous pedogenic nodules are locally common, while intact procolophonids, dicynodonts and cynodonts have been recorded from burrow infills (Groenewald and Kitching, 1995). Fragmentary rolled bone and teeth (e.g. dicynodont tusks) are found in the intraformational conglomerates at the base of some the channel sandstones

Fig. 35. Skulls of two key tetrapod genera from the Early Triassic Lystrosaurus Assemblage Zone of the Main Karoo Basin: the pig-sized dicynodont Lystrosaurus (A) and the small primitive Procolophon (B) (From Groenewald and Kitching, 1995).

Fossils from the Katberg Formation in the Middelburg sheet area are briefly reviewed by Cole et al. (2004). An unusually diverse LAZ assemblage has recently been recorded from Barendskraal near Middelburg by Damiani et al. (2003a). The spectrum of nine or more tetrapod species found here includes Lystrosaurus (albeit with low abundance), therocephalians, and several procolophonid reptiles. The poorly-preserved fossil flora recorded by Gastaldo et al. (2005) from the basal Katberg at Carlton Heights near Noupoort is of special interest because plant fossils are so rare in this stratigraphic interval.

31 John E. Almond (2013) Natura Viva cc

Fig. 36. Reconstruction of Early Triassic biotas of the Lystrosaurus Assemblage Zone (From Benton 2003 When life nearly died). Animals illustrated here include the crocodile- like archosaur reptile Proterosuchus (top) and below this the dominant, pig-sized dicyndont Lystrosaurus, a small predatory therocephalian (middle left), several small lizard- like reptiles such as procolophonids (middle right), and two large amphibians (bottom). Plants shown here include several ferns and reedy horsetails.

No skeletal remains of vertebrates (bones, teeth) were recorded within Katberg mudrocks, sandstone or parabreccias in the R61 study area during the present field survey. This may in part be due to poor exposure of the more fossiliferous mudrock facies. Few well-developed palaeosol horizons were seen, and the calcrete nodule here have generally been modified by secondary ferruginisation and silicification which may have destroyed any original fossil content (Fig. 37). The same apples to the parabreccias (Fig. 10) which elsewhere in the Katberg may contain transported fossil bones and teeth but which are often extensively baked in the study area.

32 John E. Almond (2013) Natura Viva cc Important assemblages of large vertebrate burrows within the Katberg Formation are exposed in road cuttings along the R61 at km 18.9 (Loc. 067, some 5 km WSW of the present study area) and have been described in detail by Bordy et al. (2011) (their site 2). The burrows consist of gently inclined tunnels 25 to 40 cm across with rounded terminations and a circular to elliptical cross- section. They are usually unlined and variously infilled with massive sandstone, mudrock and / or paraconglomerate indicating passive infilling (e.g. during floods). The burrows may be 3 m or more long and penetrate up to 1.5 m deep into the surrounding mudrocks and / or sandstone. Their surface may be smooth or ornamented with coarse scratch marks generated by the claws, tusks or beak of the trace-maker which may well have been a dicynodont therapsid such as Lystrosaurus. According to Bordy et al. (2011) these burrows probably functioned as temporary resting, hiding or aestivating refuges – perhaps durig intense drought episodes - rather than as permanent dwelling or breeding structures. At Loc. 067 the large inclined burrows are mainly preserved within purple- brown mudrock facies, although they also occasionally penetrate into channel sandstone tops, and in some cases are closely associated with well-developed mudcracked horzons (Figs. 38 and 39).

Irregular-shaped sandstone structures hosted by mudrock within the Katburg Formation road cutting at Loc. 080 might also be vertebrate burrow fills (Fig. 40). Mudrock horizons within the thin- bedded Katburg succession at Loc. 077 features several small, oblique to sigmoidal subcylindrical burrows that are some 2.5 cm wide with a smooth to knobbly surface (Fig. 41). These are probably attributable to small terrestrial or aquatic vertebrates living on the distal floodplain or in lakes. Small burrows (e.g. Katburgia, Macanopsis) attributed to crustaceans or terrestrial arthropods have been recorded elsewhere from damp or submerged Katburg mudrocks by previous authors (Bordy et al. 2011, Gastaldo & Rolerson 2008 and refs. therein). The burrows seen at Loc. 077 are similar in gross form and size to Katbergia which has been extensively recorded close to the Permo- Triassic boundary but further studies (e.g. presence / absence of surface scratch marks) are needed to confirm this identification.

Fig. 37. Palaeosol horizon associated with large calcrete nodules (Loc. 072). These ancient fossil soils are often associated with fossil vertebrate remains in the Beaufort Group, but appear to have often been secondarily modified (e.g. ferruginised) as a result of dolerite intrusion in the study area (Hammer = 27 cm).

33 John E. Almond (2013) Natura Viva cc

Fig. 38. Longitudinal section through a large, gently-inclined vertebrate burrow within Katburg Formation mudrocks at Loc. 067 (5 km west of study area along R61). The burrow is over 2 m long, elliptical in cross-section, and infilled with sandstone below and mudrock within the upper section.

Fig. 39. Elliptical cross-sections through large vertebrate burrows (arrowed) associated with mudcracked overbank mudrocks within the Katberg Formation (Loc. 067).The burrows here are c. 20 cm wide.

34 John E. Almond (2013) Natura Viva cc

Fig. 40. Irregular-shaped, sandstone-infilled structures embedded within a thin mudrock interbed within the Katberg Formation at Loc. 080 (Hammer = 27 cm). These structures may well be vertebrate burrow infills.

Fig. 41. Several small curved to sigmoidal invertebrate burrows (arrowed) with a calcretised silty infill – possibly Katbergia – preserved within thin-bedded Katburg mudrocks at Loc. 077 (Scale in cm).

35 John E. Almond (2013) Natura Viva cc

3.2. Fossils in the Burgersdorp Formation

The Burgersdorp Formation is characterized by a diverse continental fossil biota of Early to Mid Triassic (Olenekian to ) age, some 249 to 237 milllion years old (Kitching 1995, Rubidge 2005, Neveling et al. 2005). Karoo fossil biotas of this age are of special interest in that they document the recovery of life on land following the catastrophic end-Permian mass extinction event (Benton 2003). The Burgersdorp fauna is dominated by a wide variety of tetrapod taxa, notably a range of amphibians, reptiles and therapsids (“mammal-like reptiles”). This distinctive biota is referred to the Cynognathus Assemblage Zone (= Kannemeyeria – Diademodon Assemblage Zone of earlier authors; see Keyser & Smith 1977-78, Kitching 1995). Comparable Triassic faunas have been described from various parts of the ancient supercontinent Pangaea, including Russia, China, India, Argentina, Australia and Antarctica.

Useful accounts of the palaeontological heritage of this stratigraphic unit – which has recently been recognised as one of the richest Early-Mid Triassic biotas worldwide – are given by Kitching (1977, 1995), Keyser and Smith (1977-78), MacRae (1999), Smith et al. (1998, 2002), Hancox (2000; see also many references therein), Cole et al. (2004) and Rubidge (2005). The Burgersdorp biotas include a rich freshwater vertebrate fauna, with a range of fish groups (e.g. sharks, lungfish, coelacanths, ray-finned bony fish such as palaeoniscoids) as well as large capitosaurid and trematosuchid amphibians; the latter are of considerable important for long-range biostratigraphic correlation. The interesting reptile fauna includes lizard-like sphenodontids, beaked , and various primitive archosaurs (distant relatives of the ) such as the crocodile-like erythrosuchids, some of which reached body lengths of 5m, as well as the more gracile Euparkeria (Fig. 42). The therapsid fauna contains large herbivorous dicynodonts like Kannemeyeria (Fig. 43), which may have lived in herds, plus several small to medium-sized carnivorous or herbivorous therocephalians (e.g. Bauria) and advanced cynodonts. The most famous here is probably the powerful-jawed Cynognathus (Fig. 43), but remains of the omnivorous Diademodon are much commoner. Tetrapods are also represented by several fossil trackways while large Cruziana–like burrow systems with coarsely scratched ventral walls are attributed to burrowing vertebrates (cf Shone 1978). Locally abundant vertebrate burrows have been attributed to small procolophonid reptiles (Groenewald et al. 2001). Important new studies on lacustrine biotas in the northern Burgersdorp outcrop area have yielded rich microvertebrate faunas as well as vertebrate coprolites; sites such as Driefontein in the Free State are now among the best- documented non-marine occurrences of Early Triassic age anywhere in the world (Bender & Hancox 2003, 2004, Hancox et al. 2010, Ortiz et al. 2010 and refs. therein).

Fig. 42. Reconstruction of the small (c. 0.5m long) bipedal reptile Euparkeria, a primitive member of the archosaur group from which dinosaurs evolved later in the Triassic Period.

Contemporary invertebrate faunas are still very poorly known. Freshwater unionid molluscs are rare, while the chitinous exoskeletons of the once-abundant terrestrial arthropods do not preserve well in the highly oxidising arid-climate sediments found here; arthropod trace fossils are known but so far no fossil insects. Likewise fossil plants of the characteristic Triassic Dicroidium Flora are 36 John E. Almond (2013) Natura Viva cc poorly represented and low in diversity. They include lycophytes (club mosses), ferns (including horsetails like Schizoneura), “seed ferns” (e.g. Dicroidium) and several gymnospermous groups (conifers, ginkgos, cycads etc) (Anderson & Anderson, 1985, Bamford 2004). A small range of silicified gymnospermous fossil woods are also present including , Podocarpoxylon and Mesembrioxylon (Bamford 1999, 2004).

According to Kitching (1963, 1995) isolated, dispersed fossil bones, as well as some well- articulated skeletons, are associated with “thin localised lenses of silty sandstone” within the Burgersdorp Formation. Pedogenic, brown-weathering calcrete concretions occasionally contain complete fossil skeletons, while transported “rolled” bone is associated with intraformational conglomeratic facies at the base of channel sandstones. Fossil diversity decreases upwards through the succession. Complete tetrapod specimens are commoner lower down and amphibian remains higher up (Kitching 1995).

Fig. 43. Reconstruction of typical therapsids of the Early Triassic Cynognathus Assemblage Zone - the large tusked herbivorous dicynodont Kannemeyeria and the predatory, bear-sized cynodont Cynognathus. The inset shows the heavily-built skull of Cynognathus (c. 30cm long) in lateral view.

The biostratigraphy of the Early–Mid Triassic sediments of the Karoo Supergroup (Tarkastad Subgroup) has been the focus of considerable palaeontological research in recent years, and the subdivision of the Cynognathus Assemblage Zone into three subunits has been proposed by several authors (See Hancox et al., 1995, Hancox 2000, Neveling et al., 2005, Rubidge 2005, Abdala et al. 2005, and refs therein). Recent research has also emphasized the rapidity of faunal turnover during the transition between the sand-dominated Katberg Formation (Lystrosaurus Assemblage Zone) and the overlying mudrock-dominated Burgersdorp Formation (Neveling et al., 2005). In the proximal (southern) part of the basin the abrupt faunal turnover occurs in the uppermost sandstones of the Katberg Formation and the lowermost sandstones of the Burgersdorp Formation (ibid., p.83 and Neveling 2004). This recent work shows that the Cynognathus Assemblage Zone correlates with the entire Burgersdorp Formation; previous

37 John E. Almond (2013) Natura Viva cc authors had proposed that the lowermost Burgersdorp beds belonged to the Lystrosaurus Assemblage Zone (e.g. Keyser & Smith 1977-78, Johnson & Hiller 1990, Kitching 1995).

The Burgersdorp Formation is poorly exposed along the R61 between Cradock and Tarkastad. No vertebrate remains or other fossil material were observed within either the sandstone or mudrock facies not the associated paraconglomerates during the present study. Well-developed palaesols associated with calcrete nodules were also rarely seen. Burgersdorp fossils recorded in the Queenstown area some 60 km east of Tarkastad have been briefly reviewed by Almond (2010). Cole et al. (2004) review the Cynognathus Assemblage Zone fossils in the Middelburg sheet area and note that fossils tend to be rare and fragmentary in the lower Burgersdorp Formation (as represented here), although some well-preserved material is also known.

3.3. Fossils in the Karoo Dolerite Suite

Dolerite outcrops within the study area are in themselves of no palaeontological significance since these are high temperature igneous rocks emplaced at depth within the Earth’s crust. However, as a consequence of their proximity to large dolerite intrusions the adjacent Karoo Supergroup sediments have often been thermally metamorphosed or “baked” (i.e. recrystallised, impregnated with secondary minerals). Embedded fossil material of phosphatic composition, such as bones and teeth, was frequently altered by baking. Bones may become blackened and they can be very difficult to extract from the hard matrix by mechanical preparation. Thermal by dolerite intrusions therefore tends to reduce the palaeontological heritage potential of adjacent Beaufort Group sediments.

No vertebrate bones or teeth, or other fossil remains, were recorded within baked sediments associated with dolerite intrusions in the R61 study area.

3.4. Fossils in Late Caenozoic superficial sediments

The Karoo “drift” deposits have been comparatively neglected in palaeontological terms for the most part. However, they may occasionally contain important fossil biotas, notably the bones, teeth and horn cores of (e.g. Pleistocene mammal faunas at Florisbad, Cornelia and Erfkroon, Free State and elsewhere; Wells & Cooke 1942, Cooke 1974, Skead 1980, Klein 1984, Brink, J.S. 1987, Bousman et al. 1988, Bender & Brink 1992, Brink et al. 1995, MacRae 1999, Churchill et al. 2000 Partridge & Scott 2000) including skeletal remains of early humans (Grine et al. 2007). Other late Caenozoic fossil biotas from these superficial deposits include non-marine molluscs (bivalves, gastropods), ostrich egg shells, trace fossils (e.g. calcretised termitaria, coprolites), and plant remains such as palynomorphs in organic-rich alluvial horizons (Scott 2000) and diatoms in pan sediments. It is notable that in the sheet area to the north of the Queenstown sheet Bruce et al. (1983) report abundant plant material throughout the Quaternary alluvial deposits as well as rounded, transported Earlier Stone Age implements in the Pleistocene basal gravels.

No fossils were recorded within the superficial sediments in the R61 study area.

4. CONCLUSIONS & RECOMMENDATIONS

Section 3 of the R61 between Cradock and Tarkastad (km 24.2 to km 75) traverses fluvial sediments of the Tarkastad Subgroup (upper Beaufort Group) of Early to Middle Triassic age that are extensively intruded by dolerites of the Early Jurassic Karoo Dolerite Suite. Bedrock exposure of the Tarkastad sediments is generally poor and strongly biased towards road cuttings through more resistant-weathering sandstone packages as well as baked sediments adjacent to dolerite intrusions. Potentially fossiliferous mudrock successions are poorly represented. This applies 38 John E. Almond (2013) Natura Viva cc especially to the more recessive-weathering Burgersdorp Formation, of which very few informative exposures were seen along the study route. Some of the road cuttings through the Katberg Formation (e.g. at km 38 and 41.6) and the Burgersdorp Formation (e.g. at km 70.6) are of sedimentological interest, and this would undoubtedly apply to the future widened cuttings as well. No vertebrate or plant body fossils were observed within the Tarkastad Subgroup rocks which appear to be at most very sparsely fossiliferous along the study route. Few well-developed palaeosol (ancient soil) horizons marked by pedogenic calcrete nodules, with which vertebrate fossils are often associated, were seen here.

The only palaeontological remains recorded within the Tarkastad Subgroup bedrocks along Section 3 of the R61 were small invertebrate burrows – possibly the arthropod trace fossil Katbergia - within Katburg mudrocks at one site (km 38) as well as possible large vertebrate burrows within overbank mudrocks at km 41.6. These Katburg Formation trace fossils are quite common and of fairly low heritage significance so no special conservation measures are proposed here. Good examples of large, shallowly inclined, sand- or mud-infilled vertebrate burrows excavated into the Katberg mudrocks are well seen in a road cutting close to Cradock (km 18.9), but this important, well-known locality lies some 5 km west of the present study area.

Late Caenozoic gravelly, silty and sandy alluvial deposits observed within river and stream banks en route are generally of low palaeontological sensitivity and no fossil or subfossil material was recorded therein.

All of the five proposed borrow pit sites (BP1-BP5) as well as the proposed new hard rock quarry (HRQ1) are underlain by Karoo dolerite intrusions and are of no palaeontological heritage significance. The dolerite in some cases is deeply weathered to yield resistant, rounded corestones embedded in friable sabunga. Adjacent sedimentary country rocks have been baked to quartzites and hornfels, seriously compromising their fossil heritage potential. No fossils were observed within these thermally metamorphosed country rocks.

It is concluded that the proposed upgrade of Section 3 of the R61 between Cradock and Tarkastad - including the proposed borrow pit and quarry excavations as well as modifications to stormwater structures - is of LOW palaeontological heritage significance. Pending the discovery of substantial new fossils during before or during development, no further specialist studies or mitigation in this respect are considered necessary for this road project.

Should substantial fossil remains be exposed during construction, however, such as vertebrate bones and teeth, plant-rich fossil lenses or dense fossil burrow assemblages, the Environmental Control Officer should safeguard these, preferably in situ, and alert ECPHRA (i.e. The Eastern Cape Provincial Heritage Resources Authority. Contact details: Mr Sello Mokhanya, 74 Alexander Road, King Williams Town 5600; [email protected]) as soon as possible so that appropriate action (e.g. recording, sampling or collection) can be taken by a professional palaeontologist. These recommendations should be incorporated into the Environmental Management Plan for the road project.

6. ACKNOWLEDGEMENTS

Dr Norbert Klages of Arcus GIBB Engineering and Science, Port Elizabeth, is thanked for commissioning this specialist study and for kindly providing extensive background information to assist with the fieldwork.

39 John E. Almond (2013) Natura Viva cc 7. KEY REFERENCES

ABDALA, F., HANCOX, P.J. & NEVELING, J. 2005. Cynodonts from the uppermost Burgersdorp Formation, South Africa, and their bearing on the biostratigraphy and correlation of the Triassic Cynognathus Assemblage Zone. Journal of Vertebrate Paleontology 25, 192-199.

ABDALA, F., CISNEROS, J.C. & SMITH, R.M.H. 2006. Faunal aggregation in the Early Triassic Karoo Basin: earliest evidence of shelter-sharing behaviour among tetrapods. Palaios 21, 507- 512.

ALMOND, J.E. 2010. Proposed iNca Energy wind facility on the farms Cathcarts Gift 311 and Latham 205 near Queenstown, Eastern . Palaeontological impact assessment: combined desktop and scoping study, 30 pp. Natura Viva cc, Cape Town.

ALMOND, J.E. 2011. Proposed Mainstream wind farm near Noupoort, Pixley ka Seme District Municipality, Northern Cape Province. Palaeontological desktop study, 20 pp. Natura Viva cc, Cape Town.

ALMOND, J.E., DE KLERK, W.J. & GESS, R. 2008. Palaeontological heritage of the Eastern Cape. Draft report for SAHRA, 30 pp. Natura Viva cc, Cape Town.

ANDERSON, J.M. & ANDERSON, H.M. 1985. Palaeoflora of southern Africa. Prodromus of South African megafloras, Devonian to Lower , 423 pp. Botanical Research Institute, Pretoria & Balkema, Rotterdam.

BAMFORD, M. 1999. Permo-Triassic fossil woods from the South African Karoo Basin. Palaeontologia africana 35, 25-40.

BAMFORD, M.K. 2004. Diversity of woody vegetation of Gondwanan southern Africa. Gondwana Research 7, 153-164.

BENDER, P.A. & BRINK, J.S. 1992. A preliminary report on new large mammal fossil finds from the Cornelia

BENDER, P.A. & HANCOX, P.J. 2003. Fossil fishes of the Lystrosaurus and Cynognathus Assemblage Zones, Beaufort Group, South Africa: correlative implications. Council for Geoscience, Pretoria, Bullletin 136, 1-27.

BENDER, P.A. & HANCOX, P.J. 2004. Newly discovered fish faunas from the Early Triassic, Karoo Basin, South Africa, and their correlative implications. Gondwana Research 7, 185-192.

BENTON, M.J. 2003. When life nearly died. The greatest mass extinction of them all, 336 pp. Thames & Hudson, London.

BOK, S.N. 2011. Four potential wind farm sites near Lady Grey, Noupoort, Prieska and Louriesfontein. Geotechnical desktop study, 18 pp. Jeffares & Green (Pty) Ltd.

BORDY, E. M., SZTANÓ, O., RUBIDGE, B.S. AND BUMBY, A. 2009. Tetrapod burrows in the southwestern main Karoo Basin (Lower Katberg Formation, Beaufort Group), South Africa. Extended Abstracts of the 15th Biennial Conference of the Palaeontological Society of Southern Africa. September 11-14, Matjiesfontein, South Africa. Palaeontologia Africana 44, 95-99.

BORDY, E.M., SZTANÓ, O, RUBDIGE, B. & BUMBY, A. 2011. Early Triassic vertebrate burrows from the Katberg Formation of the south-western Karoo Basin, South Africa. Lethaia 44, 33-45.

40 John E. Almond (2013) Natura Viva cc BOTHA, J. & SMITH, R.M.H. 2007. Lystrosaurus species composition across the Permo-Triassic boundary in the Karoo Basin of South Africa. Lethaia 40, 125-137.

BOUSMAN, C.B. et al. 1988. Palaeoenvironmental implications of Late Pleistocene and Holocene valley fills in Blydefontein Basin, Noupoort, C.P., South Africa. Palaeoecology of Africa 19: 43-67.

BRINK, J.S. 1987. The archaeozoology of Florisbad, Orange Free State. Memoirs van die Nasionale Museum 24, 151 pp.

BRINK, J.S. et al. 1995. A new find of Megalotragus priscus (Alcephalini, Bovidae) from the Central Karoo, South Africa. Palaeontologia africana 32: 17-22.

BRUCE, R.W., KRUGER, G.P. & JOHNSON, M.R. 1983. Die geologie van die gebied Aliwal- Noord. Explanation to 1: 250 000 geology Sheet 3026 Aliwal-Noord, 7 pp. Council for Geoscience, Pretoria.

CHURCHILL, S.E. et al. 2000. Erfkroon: a new Florisian fossil locality from fluvial contexts in the western Free State, South Africa. South African Journal of Science 96: 161-163.

CLUVER, M.A. 1978. Fossil reptiles of the South African Karoo. 54pp. South African Museum, Cape Town.

COLE, D.I., NEVELING, J., HATTINGH, J., CHEVALLIER, L.P., REDDERING, J.S.V. & BENDER, P.A. 2004. The geology of the Middelburg area. Explanation to 1: 250 000 geology Sheet 3124 Middelburg, 44 pp. Council for Geoscience, Pretoria.

COOKE, H.B.S. 1974. The fossil mammals of Cornelia, O.F.S., South Africa. In: Butzer, K.W., Clark, J.D. & Cooke, H.B.S. (Eds.) The geology, archaeology and fossil mammals of the Cornelia Beds

DAMIANI, R., NEVELING, J., MODESTO, S. & YATES, A. 2003a. Barendskraal, a diverse amniote locality from the Lystrosaurus Assemblage Zone, Early Triassic of South Africa. Palaeontologia Africana 39, 53-62.

DAMIANI, R., MODESTO, S., YATES, A. & NEVELING, J. 2003b. Earliest evidence for cynodont burrowing. Proceedings of the Royal Society of London B. 270, 1747-1751.

DINGLE, R.V., SIESSER, W.G. & NEWTON, A.R. 1983. and Tertiary geology of southern Africa. viii + 375 pp. Balkema, Rotterdam.

DUNCAN, A.R. & MARSH, J.S. 2006. The Karoo Igneous Province. In: Johnson, M.R., Anhaeusser, C.R. & Thomas, R.J. (Eds.) The , pp. 501-520. Geological Society of South Africa, Marshalltown.

GASTALDO, R.A., ADENDORFF, R., BAMFORD, M., LABANDEIRA, C.C., NEVELING, J. & SIMS, H. 2005. Taphonomic trends of macrofloral assemblages across the Permian – Triassic boundary, Karoo Basin, South Africa. Palaios 20, 479-497.

GASTALDO, R.A. & ROLERSON, M.W. 2008. Katbergia Gen. Nov., a new trace fossil from the Upper Permian and Lower Triassic rocks of the Karoo Basin: implications for palaeoenvironmental conditions at the P/TR extinction event. Palaeontology 51, 215-229.

GRINE, F.E., BAILEY, R.M., HARVATI, K., NATHAN, R.P., MORRIS, A.G., HENDERSON, G.M., RIBOT, I., PIKE, A.W. 2007. Late Pleistocene human skull from Hofmeyr, South Africa, and modern human origins. Science 315, 226–9.

41 John E. Almond (2013) Natura Viva cc GROENEWALD, G.H. 1991. Burrow casts from the Lystrosaurus-Procolophon Assemblage-zone, Karoo Sequence, South Africa. Koedoe 34, 13-22.

GROENEWALD, G.H. 1996. Stratigraphy of the Tarkastad Subgroup, Karoo Supergroup, South Africa. Unpublished PhD thesis, University of Port Elizabeth, South Africa.

GROENEWALD, G.H. & KITCHING, J.W. 1995. Biostratigraphy of the Lystrosaurus Assemblage Zone. Pp. 35-39 in RUBIDGE, B.S. (ed.) Biostratigraphy of the Beaufort Group (Karoo Supergroup). South African Committee for Stratigraphy, Biostratigraphic Series No. 1, 46 pp. Council for Geoscience, Pretoria.

GROENEWALD, G. H., J. WELMAN, AND J. A. MACEACHERN. 2001. Vertebrate burrow complexes from the Early Triassic Cynognathus Assemblage Zone (Driekoppen Formation, Beaufort Group) of the Karoo Basin, South Africa. Palaios16,148–160.

HANCOX, P.J. 2000. The continental Triassic of South Africa. Zentralblatt für Geologie und Paläontologie Teil 1,1998, Heft 11-12, 1285-1324.

HANCOX, P.J., SHISHKIN, M.A., RUBIDGE, B.S. & KITCHING, J.W. 1995. A threefold subdivision of the Cynognathus Assemblage Zone (Beaufort Group, South Africa) and its palaeogeographic implications. South African Journal of Science 91, 143-144.

HANCOX, J., NEVELING, J. & RUBIDGE, B. 2010. Life in an Early Triassic lake: new developments at the Driefontein site, Burgersdorp Formation (Cynognathus Assemblage Zone), South Africa. Proceedings of the 16th Conference of the PSSA, Howick, Umgeni Valley Nature Reserve, 41-42.

HAYCOCK, C.A., MASON, T.R. & WATKEYS, M.K. 1994. Early Triassic palaeoenvironments in the eastern Karoo foreland basin, South Africa. Journal of African Earth Sciences 24, 79-94.

HILLER, N. & STAVRAKIS, N. 1980. Distal alluvial fan deposits in the Beaufort Group of the Eastern Cape Province. Transactions of the Geological Society of South Africa 83, 353-360.

HILLER, N. & STAVRAKIS, N. 1984. Permo-Triassic fluvial systems in the southeastern Karoo Basin, South Africa. Palaeogeography, Palaeoclimatology, Palaeoecology 34, 1-21.

JOHNSON, M.R. 1966. The stratigraphy of the Cape and Karoo Systems in the Eastern Cape Province. Unpublished MSc Thesis, Rhodes University, Grahamstown.

JOHNSON, M.R. 1976. Stratigraphy and sedimentology of the Cape and Karoo sequences in the Eastern Cape Province. Unpublished PhD thesis, Rhodes University, Grahamstown, xiv + 335 pp, 1pl.

JOHNSON, M.R. 1984. The geology of the Queenstown area. Explanation to 1: 250 000 geology Sheet 3126 Queenstown, 21 pp. Council for Geoscience, Pretoria.

JOHNSON, M.R. & HILLER, N. 1990. Burgersdorp Formation. South African Committee for Stratigraphy, Catalogue of South African Lithostratigraphic Units 2, 9-10. Council for Geoscience, Pretoria.

JOHNSON, M.R., VAN VUUREN, C.J., VISSER, J.N.J., COLE, D.I., DE V. WICKENS, H., CHRISTIE, A.D.M., ROBERTS, D.L. & BRANDL, G. 2006. Sedimentary rocks of the Karoo Supergroup. In: Johnson, M.R., Anhaeusser, C.R. & Thomas, R.J. (Eds.) The geology of South Africa, pp. 461-499. Geological Society of South Africa, Marshalltown.

42 John E. Almond (2013) Natura Viva cc KARPETA, W.P. & JOHNSON, M.R. 1979. The geology of the Umtata area. Explanation to 1: 250 000 geology Sheet 3128 Umtata, 16 pp. Council for Geoscience, Pretoria.

KEYSER, A.W. & SMITH, R.M.H. 1977-78. Vertebrate biozonation of the Beaufort Group with special reference to the Western Karoo Basin. Annals of the Geological Survey of South Africa 12: 1-36.

KITCHING, J.W. 1963. Notes on some fossil pockets and bone beds in the Cynognathus-Zone in the Burghersdorp and Lady Frere Districts. Palaeontologia Africana 8, 113-118.

KITCHING, J.W. 1977. The distribution of the Karroo vertebrate fauna, with special reference to certain genera and the bearing of this distribution on the zoning of the Beaufort beds. Memoirs of the Bernard Price Institute for Palaeontological Research, University of the , No. 1, 133 pp (incl. 15 pls).

KITCHING, J.W. 1995. Biostratigraphy of the Cynognathus Assemblage Zone. Pp. 13-17 in Rubidge, B.S. (ed.) Biostratigraphy of the Beaufort Group (Karoo Supergroup). South African Committee for Stratigraphy, Biostratigraphic Series No. 1. Council for Geoscience, Pretoria.

KLEIN, R.G. 1984. The large mammals of southern Africa: Late Pliocene to Recent. In: Klein, R.G. (Ed.) Southern African prehistory and paleoenvironments, pp 107-146. Balkema, Rotterdam.

MACRAE, C. 1999. Life etched in stone. Fossils of South Africa. 305pp. The Geological Society of South Africa, .

MCCARTHY, T. & RUBIDGE, B. 2005. The story of Earth and life: a southern African perspective on a 4.6-billion-year journey. 334pp. Struik, Cape Town.

MODESTO, S.P. & BOTHA-BRINK, J. 2010. A burrow cast with Lystrosaurus skeletal remains from the Lower Triassic of South Africa. Palaios 25, 274-281.

NEVELING, J., RUBIDGE, B.S. & HANCOX, P.J. 1999. A lower Cynognathus Assemblage Zone fossil from the Katberg Formation (Beaufort Group, South Africa). South African Journal of Science 95, 555-556.

NEVELING, J. 2004. Stratigraphic and sedimentological investigation of the contact between the Lystrosaurus and the Cynognathus Assemblage Zones (Beaufort Group: Karoo Supergroup). Council for Geoscience, Pretoria, Bulletin, 137, 164pp.

NEVELING, J., HANCOX, P.J. & RUBIDGE, B.S. 2005. Biostratigraphy of the lower Burgersdorp Formation (Beaufort Group; Karoo Supergroup) of South Africa – implications for the stratigraphic ranges of early Triassic tetrapods. Palaeontologia Africana 41, 81-87.

ORTIZ, D., LEWIS, P.J., KENNEDY, A.M., BHULLAR, B.S. & HANCOX, J. 2010. Preliminary analysis of lungfish (Dipnoi) tooth plates from Driefontein, South Africa. Proceedings of the 16th Conference of the PSSA, Howick, Umgeni Valley Nature Reserve, 72-74.

PARTRIDGE, T.C. & SCOTT, L. 2000. Lakes and pans. In: Partridge, T.C. & Maud, R.R. (Eds.) The Cenozoic of southern Africa, pp.145-161. Oxford University Press, Oxford.

PARTRIDGE, T.C., BOTHA, G.A. & HADDON, I.G. 2006. Cenozoic deposits of the interior. In: Johnson, M.R., Anhaeusser, C.R. & Thomas, R.J. (Eds.) The geology of South Africa, pp. 585-604. Geological Society of South Africa, Marshalltown.

RETALLACK, G.J., SMITH, R.M.H. & WARD, P.D. 2003. Vertebrate extinction across the Permian-Triassic boundary in the Karoo Basin, South Africa. Geological Society of America Bulletin 115, 1133-1152. 43 John E. Almond (2013) Natura Viva cc

RETALLACK, G.J., METZGER, C.A., GREAVER, T., HOPE JAHREN, A., SMITH, R.M.H. & SHELDON, N.D. 2006. Middle – Late Permian mass extinction on land. GSA Bulletin 118, 1398- 1411.

RUBIDGE, B.S. (Ed.) 1995. Biostratigraphy of the Beaufort Group (Karoo Supergroup). South African Committee for Biostratigraphy, Biostratigraphic Series No. 1., 46 pp. Council for Geoscience, Pretoria.

RUBIDGE, B.S. 2005. Re-uniting lost continents – fossil reptiles from the ancient Karoo and their wanderlust. South African Journal of Geology 108: 135-172.

SAHRA 2013. Minimum standards: palaeontological component of heritage impact assessment reports, 15 pp. South African Heritage Resources Agency, Cape Town.

SHONE, R.W. 1978. Giant Cruziana from the Beaufort Group. Transactions of the Geological Society of South Africa 81, 327-329.

SKEAD, C.J. 1980. Historical mammal incidence in the Cape Province. Volume 1: The Western and Northern Cape, 903pp. Department of Nature and Environmental Conservation, Cape Town.

SMITH, R. M.H., TURNER, B.R., HANCOX, P.J., RUBDIGE, B.R. & CATUNEANU, O. 1998. Trans-Karoo II: 100 million years of changing terrestrial environments in the main Karoo Basin. Guidebook Gondwana-10 International Conference, University of Cape Town, South Africa, 117 pp.

SMITH, R.H.M. & WARD, P.D. 2001. Pattern of vertebrate extinction across an event bed at the Permian-Triassic boundary in the Karoo Basin of South Africa. Geology 29, 1147-1150.

SMITH, R.M.H., HANCOX, P.J., RUBIDGE, B.S., TURNER, B.R. & CATUNEANU, O. 2002. Mesozoic ecosystems of the Main Karoo Basin: from humid braid plains to arid sand sea. Guidebook 8th International Symposium on Mesozoic Terrestrial Ecosystems, Cape Town, South Africa, 116 pp.

SMITH, R. & BOTHA, J. 2005. The recovery of terrestrial vertebrate diversity in the South African Karoo Basin after the end-Permian extinction. Comptes Rendus Palevol 4, 555-568.

STAVRAKIS, N. 1980. of the Katberg Sandstone and adjacent formations in the south-eastern Karoo Basin. Transactions of the Geological Society of South Africa 83, 361-374.

VIGLIETTI, P. 2010. Origin, sedimentology and taphonomy of an Early Triassic Lystrosaurus bonebed, Katberg Formation, Karoo Basin, South Africa. Proceedings of the 16th Conference of the Palaeontological Society of Southern Africa, Howick, August 5-8, 111a-111c.

VISSER, J.N.J. & DUKAS, B.A. 1979. Upward-fining fluviatile megacycles within the Beaufort Group, north of Graaff-Reinet, Cape Province. Transactions of the Geological Society of South Africa 82, 149-154.

VISSER, D.J.L. et al. 1989. The geology of the Republics of South Africa, Transkei, Bophuthatswana, Venda and Ciskei and the Kingdoms of Lesotho and Swaziland. Explanation of the 1: 1 000 000 geological map, fourth edition, 491 pp. Council for Geoscience, Pretoria.

WARD, P.D., BOTHA, J., BUICK, R., DE KOCK, M.O., ERWIN, D.H., GARRISON, G.H., KIRSCHVINK, J.L. & SMITH, R.M.H. 2005. Abrupt and gradual extinction among Late Permian land vertebrates in the Karoo Basin, South Africa. Science 307, 709-714.

44 John E. Almond (2013) Natura Viva cc WELLS, L.H. & COOKE, H.B.S. 1942. The associated fauna and culture of Vlakkraal thermal springs, O.F.S.; III, the faunal remains. Transactions of the Royal Society of South Africa 29: 214- 232.

WELMAN, J., GROENEWALD, G.H. & Kitching, j.w. 1991. Confirmation of the occurrence of Cynognathus Zone (Kannemeyeria – Diademodon Assemblage-Zone) deposits (uppermost Beaufort Group) in the northeastern Orange Free State, South Africa. South African Journal of Geology 94, 245-248.

8. QUALIFICATIONS & EXPERIENCE OF THE AUTHOR

Dr John Almond has an Honours Degree in Natural Sciences (Zoology) as well as a PhD in Palaeontology from the University of Cambridge, UK. He has been awarded post-doctoral research fellowships at Cambridge University and in Germany, and has carried out palaeontological research in Europe, North America, the Middle East as well as North and South Africa. For eight years he was a scientific officer (palaeontologist) for the Geological Survey / Council for Geoscience in the RSA. His current palaeontological research focuses on fossil record of the - boundary and the Cape Supergroup of South Africa. He has recently written palaeontological reviews for several 1: 250 000 geological maps published by the Council for Geoscience and has contributed educational material on fossils and evolution for new school textbooks in the RSA.

Since 2002 Dr Almond has also carried out palaeontological impact assessments for developments and conservation areas in the Western, Eastern and Northern Cape under the aegis of his Cape Town-based company Natura Viva cc. He is a long-standing member of the Archaeology, Palaeontology and Meteorites Committee for Heritage (HWC) and an advisor on palaeontological conservation and management issues for the Palaeontological Society of South Africa (PSSA), HWC and SAHRA. He is currently compiling technical reports on the provincial palaeontological heritage of Western, Northern and Eastern Cape for SAHRA and HWC. Dr Almond is an accredited member of PSSA and APHP (Association of Professional Heritage Practitioners – Western Cape).

Declaration of Independence

I, John E. Almond, declare that I am an independent consultant and have no business, financial, personal or other interest in the proposed project, application or appeal in respect of which I was appointed other than fair remuneration for work performed in connection with the activity, application or appeal. There are no circumstances that compromise the objectivity of my performing such work.

Dr John E. Almond Palaeontologist Natura Viva cc

45 John E. Almond (2013) Natura Viva cc Appendix: GPS LOCALITY DATA

All GPS readings were taken in the field using a hand-held Garmin GPSmap 60CSx instrument. The datum used is WGS 84.

19 to 20 December 2012 – R61 (Section 3) project

Location number South East 067 S32 06 15.0 E25 47 37.2 068 S32 06 13.9 E25 47 32.3 069 S32 05 00.2 E25 50 37.8 070 S32 05 02.0 E25 50 31.6 071 S32 04 14.1 E25 51 22.8 072 S32 03 34.2 E25 52 03.4 073 S32 03 17.1 E25 52 35.3 074 S32 02 10.0 E25 54 51.2 075 S32 01 08.8 E25 57 22.8 076 S32 00 48.3 E25 57 48.3 077 S32 00 43.3 E25 57 52.5 078 S32 00 20.9 E25 57 55.0 079 S31 59 58.6 E25 59 03.1 080 S31 59 40.2 E25 59 11.1 081 S31 59 06.9 E25 59 01.4 082 S31 57 42.0 E26 00 14.4 083 S31 55 56.2 E26 04 20.8 084 S31 55 57.5 E26 04 49.3 085 S31 55 42.8 E26 07 29.1 086 S31 57 02.5 E26 11 10.1 087 S31 57 39.6 E26 12 23.1 088 S31 59 22.4 E26 13 16.0 089 S31 59 08.2 E26 13 00.2 090 S32 00 17.5 E25 57 45.8 091 S32 00 15.0 E25 57 46.1 092 S32 00 18.0 E25 57 50.9

46 John E. Almond (2013) Natura Viva cc