South African Journal of Botany 95 (2014) 135145

Contents lists available at ScienceDirect

South African Journal of Botany

j ourna l homepage: www.e lsev ie r .com/ locate /sa jbThe Peninsula Shale Renosterveld of Devil's Peak, Western Cape: A studyinto the vegetation and seedbank with a view towardpotential restorationO.S. Cowan a,, P.M.L. Anderson b

a Department of Environmental and Geographical Sciences, University of Cape Town, South Africab African Centre for Cities, Environmental and Geographical Science Department, University of Cape Town, South Africa Corresponding author at: Department of EnvironmenUniversity of Cape Town, Rondebosch, Cape Town 7216503798.

E-mail address: cwnoli001@gmail.com (O.S. Cowan).

http://dx.doi.org/10.1016/j.sajb.2014.09.0030254-6299/ 2014 SAAB. Published by Elsevier B.V. All riga b s t r a c ta r t i c l e i n f oArticle history:Received 23 May 2014Received in revised form 29 August 2014Accepted 6 September 2014Available online xxxx

Edited by RM Cowling

Keywords:RenosterveldRestorationVegetation surveySeedbank analysisUrban ecologyPeninsula Shale Renosterveld is a poorly understood and critically threatened vegetation type. Less than 13% of itsoriginal extent remains and the surviving fragments are restricted to the area surrounding the Cape Town citybowl. A substantial portion of Peninsula Shale Renosterveld is situated in theGroote Schuur Estate, currently con-servedwithin the TableMountain National Park. Due to a history of diverse land-use impactswhat remains todayis a complex mosaic of alien invasive species and indigenous vegetation in various states of degradation and toimprove the conservation value restoration is required. However, it is imperative that prior to restoration initia-tion data is collected from the site. To enhance the understanding of the ecosystem drivers of Peninsula ShaleRenosterveld with a view to evaluate restoration potential and inform future restoration initiatives a detailedanalysis of four typical vegetation states found within the study area was undertaken. This was accomplishedthrough a vegetation survey investigating different states of the contemporary condition of the study areaand an examination of the health of the seedbank through a burning and greenhouse experiment to ascertainwhether it could prove a useful asset in future restoration initiatives. The results show how historical drivershave created a novel ecosystem with vegetation states ranging from relatively healthy Renosterveld vegetation,indigenous vegetation requiring intervention to maintain its integrity, and regions of the study area which aredominated by alien grasses and Pinus plantations. The seedbank diversity was found to be poor, dominated byalien ephemeral species and unlikely to assist in restoration efforts. Although certain areas may retain their eco-logical integrity through passive restoration, large swathes of the study area require active restoration to returnthe degraded vegetation to functional Peninsula Shale Renosterveld.

2014 SAAB. Published by Elsevier B.V. All rights reserved.1. Introduction

Renosterveld, a highly threatened Mediterranean-type shrublandin the Cape Floristic Region of South Africa (Rebelo et al., 2006), is oneof the most species-rich assemblages of geophytes worldwide(Proches et al., 2006) and contains one third of the endemics of thegreater Cape Floristic Region (Myers et al., 2000; Rebelo et al., 2006).Renosterveld is found on comparatively nutrient-rich soils and as a re-sult it has undergone large-scale habitat transformation, predominantlydue to conversion to agriculture, in the centuries following Europeansettlement in the Cape (Kemper et al., 1999). In addition, urbanizationand afforestation have transformed and further degraded Renosterveld(Richardson et al., 1994). Today only 8% of the original extent remains(Rebelo et al., 2006).tal and Geographical Sciences,701, South Africa. Tel.: +27

hts reserved.Peninsula Shale Renosterveld, the focus of this paper, is one of the 29types of Renosterveld (Rebelo et al., 2006). Over 87%of the original 2375hectares of Peninsula Shale Renosterveld has been transformed (Rebeloet al., 2011) and the remaining patches are situated on the fringes ofthe urban centre of the city of Cape Town and as a result have a com-plex land-use history. The need to safeguard and improve the con-servation status of these fragments in the region, in particulargiven the constrained opportunities for conservation in the face ofurban expansion (Rouget et al., 2003) and predicted climate change(Hannah et al., 2007), is clear. Despite the fact that the majority of theremaining vegetation falls within currently conserved areas, much ofthis is degraded in nature and the conservation value varies consider-ably within the remnant patches. As matters stand, restoration, whileviewed as a high risk option (Crookes et al., 2013), is the singlemost im-portant tool to enhancing the conservation status of this threatened andpoorly understood vegetation type.

Tomaximize the success andminimize the cost of restoration efforts,the literature is explicit that restoration projects involve effectiveplanning (Wyant et al., 1995; Hobbs and Norton, 1996). In addition to

http://crossmark.crossref.org/dialog/?doi=10.1016/j.sajb.2014.09.003&domain=pdfhttp://dx.doi.org/10.1016/j.sajb.2014.09.003mailto:cwnoli001@gmail.comUnlabelled imagehttp://dx.doi.org/10.1016/j.sajb.2014.09.003Unlabelled imagehttp://www.sciencedirect.com/science/journal/02546299www.elsevier.com/locate/sajb

136 O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145acquiring a detailed land-use history of the study area in order to under-stand drivers of change (Szabo, 2010), a fundamental constituent of res-toration planning is the ecological description of the site designated forrestoration (Hobbs and Harris, 2001; Maestre et al., 2006). The acquisi-tion of this knowledge provides perspective and aids in the determina-tion of physical and ecological design specifications for restorationinitiatives (Society for Ecological Restoration International Policy andWorking Group, 2004). Furthermore, the availability of this data is a ne-cessity if one is to evaluate changes as a result of restoration efforts(Wissmar and Beschta, 1998). Ideally, the study area description shouldprovide extensive lists of fauna and flora (including their conservationstatus); a description of seral or vegetation development; an under-standing of community structural diversity (Diaz and Cadibo, 2001)and include a description of habitat requirements for keystone species(Clewell and Rieger, 1997). In fire-driven systems burning is one ofthose key habitat requirements, and here, in light of this, considerationsmust extend to the health and potential of the soil-stored seedbank toaid in restoration (Bakker et al., 1996). Although there is a dearth ofdata for Renosterveld seedbanks, in the adjacent Fynbos vegetationindigenous soil-stored seedbanks have been shown to persist in alieninvaded sites (Holmes and Cowling, 1997) and are considered an im-portant factor in restoration success (Holmes, 2002). While under ex-plored in Renosterveld, the explicit role of fire as an important toolhas been demonstrated in Fynbos in stimulating seedbanks (Holmeset al., 2000; Holmes, 2005) through the effect of heat, smoke andother fire-induced cues such as fluctuating soil temperature, elevatedoxygen levels and heat pulses (Brown, 1993; Cowling et al., 1997).

The need for restoration to improve the conservation status of Pen-insula Shale Renosterveld and the acknowledgment of the importanceof collecting preliminary site data informed the aims of the project. Tothis end, we set out to achieve the following:

(1) Develop an understanding of four pre-defined vegetation statesin a remnant patch of Peninsula Shale Renosterveld to providea current ecological description and plant community data ofthe remnant patch and to understand these as the end-pointsof the known historical drivers, and

(2) To establish the health and availability of the soil-storedseedbank and evaluate the use of fire in this respect for use infuture restoration initiatives.

2. Materials and methods

2.1. Site description

The study area (~100 ha; centered at 335650.51S; 182729.59E)falls within theWestern Cape Province of South Africa, and is specifical-ly located within the Groote Schuur Estate, an area recently incorporat-ed into the Table Mountain National Park, and situated on the northernslopes of Devil's Peak. The region's climate is Mediterranean and typi-fied by a marked peak in precipitation (mean annual precipitation720 mm) in the mild winter (mean minimum July temperature7.8 C) and summer drought (mean maximum February temperature26.7 C) (Rebelo et al., 2006). Although the Estate contains within ittwo main natural vegetation types, the study area is situated in thenorth of the Estate where the remnant patch of Peninsula ShaleRenosterveld is located (Fig. 1). The Estate is bordered by an urbanfringe, the eastern and northern slopes of Table Mountain, and theUniversity of Cape Town's upper campus. While broadly attributed asPeninsula Shale Renosterveld (Rebelo et al., 2006), the vegetation com-munity at thefine scale of this study is currently a complexmosaic of in-digenous and alien vegetation and a preliminary analysis of thevegetation of the study area (Euston-Brown, 2011) revealed four dis-tinct vegetation states present (Fig. 2): Old Renosterveld, where firehad been absent for an extended period of time; Lightly DegradedRenosterveld which last burnt in 2010; Alien Pine stands some ofwhich are currently in the process of being clear-felled; and an areacalled the Game Camp which is a grassy area that has been managedas a grazing paddock for a variety of indigenous and alien ungulates,at varying densities, in the past century. These four clearly identifiedvegetation states are characteristic of the Peninsula Shale Renosterveldwithin the study area, were deemed most likely to elucidate informa-tion around vegetation state shifts, would inform useful managementunits, and were therefore considered most relevant in inferring ecolog-ical understanding and informing future restoration efforts and in-formed the selection and location for more detailed analysis.

2.2. Site history

The study area is adjacent to the initial settlement set up by the firstEuropean settlers in 1652 and as the colony grew the land would havebeen utilized for a variety of purposes. The lowermore accessible slopeswould have been grazed and cultivated by the first free burghers whilesubsequent centuries saw the upper slopes afforested, predominantlywith pine species, to provide timber for the colony. The continuationof the commercial plantation was deemed untenable in 1962 and this,coupledwith an increase in support for conservation of indigenous veg-etation, resulted in the City beginning to clear-fell forests and by 1980most of the trees had been removed (Shaughnessy, 1980) althoughpockets of the non-invasive Pinus pinea are still present. These havebeen intentionally retained as part of a cultural landscape as a directiveof Rhodes' will which required that the Estate be maintained asEuropean-styled parkland (Combrink et al., 1992). The natural fire re-gime of the Peninsula Shale Renosterveld would also have been alteredby the arrival of the Europeans. Initially fire frequency would have beenincreased due to both accidental human ignitions and poor attempts toemulate the burning practices of the original inhabitants of the Cape(Anderson and O' Farrell, 2012), but once homesteads, farms and plan-tations had been established fire prevention would have been practicedreducing appropriatefire intervals (vanWilgen et al., 2012). The currentmanagement policy reflects the understanding of the need for con-trolled burning and allowing wildfires to burn where they will notpose an ecological threat; however it acknowledges that the priority issafety for people and property (Forsyth and van Wilgen, 2008).

The history of the study area is somewhat different to the rest ofTable Mountain because the Estate was purchased by Cecil John Rhodes,erstwhile primeminister of the Cape Colony, in 1893whereupon he initi-ated his own management plans. It was Rhodes' intention to create aparkland after the European style with open grassland interspersedwith clumps of pine trees easily accessible to the public.Within this park-land paddocks were created for a variety of animals, both exotic and in-digenous, including grazing antelope. After Rhodes' death in 1902 theimpetus behind the maintenance of the paddocks gradually waned untilwhat remainedwere only the enclosed Game Camp and the surroundingmatrix of parkland and degraded indigenous vegetation managed by theDepartment of Public Works. In accordance with Rhodes' Will the Estateremained open to the public and maintained its parkland nature. In1999, the estate was integrated into Table Mountain National Park but itretains its status as a heritage landscape (Chittenden et al., 2002). Upuntil the latter stages of the twentieth century the Game Camp washome to a herd of Wildebeest (Connochaetes gnu) and an assortment ofother ungulates; however the majority of the animals were relocatedfrom the area and today the enclosure is home to five Zebra (Equusquagga). Plans have been drawn up to explore the possibility of ex-panding the Game Camp (Coetzee, 2001); however at the time of writingthey had not been implemented.

2.3. Data collection

2.3.1. Plant community surveysIn each of the four pre-defined vegetation states, seven 100 m2

relevs were randomly selected and demarcated. Sampling was carried

Fig. 1.An image of Groote Schuur Estate showing the position of the Renosterveld and Fynbos (Mucina and Rutherford, 2006), and the location of the four vegetation stateswithin it. OR=Old Renosterveld; LD = Lightly Degraded Renosterveld; AP = Alien Pine stand; GC = Game Camp enclosure.

137O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145out in the spring of 2011. All plant species were recorded. Those notidentified on site were collected for later identification in the BolusHerbarium. In a few instances species could not be identified becausematerial was too small, or not flowering. In these instances they werestill treated as unique species, and assigned a growth form. For eachrelev species percentage cover was estimated. In the Alien Pinestand, Pinus cover was measured as percentage of trunk and saplingground cover rather than over-head canopy cover asmanymature indi-viduals did not have foliage. Once a complete species list was compiled,various literature sources were consulted to assign each species to aheight class, functional type (adapted from Cowling et al., 1994;Mucina and Rutherford, 2006), and to allocate threat status accordingto the Red List of South African Plants (SANBI, 2012). Large shrubs/trees were classified as species which grow higher than 2 m; mid-highShrubs grow to a height between 1 and 2 m, while low shrubs areshrubs smaller than 1 m in height. Forbs and Graminoids (includinggrasses, sedges and restioids)were both divided according to life history(annual or perennial), while species which grew from undergroundbulbs or corms were designated as Geophytes.

At each relev abiotic factors were also measured. Total bare ground,soil depth (the mean of 3 readings from a soil depth probe), elevationand slopewere recorded. At each relev a composite soil sample (approx-imately 500 g) was taken from four random locations at depths up to15 cm. The following soil characteristicsweremeasured in the laboratory:texture, stone volume (%), soil moisture (%), pH (KCl; see McLean, 1982),electrical resistance (ohms; see Richards, 1954), hydrogen (cmol(+)/kg),phosphorous (Bray II) (mg/kg; see Bray and Kurtz, 1945), potassium(mg/kg), organic carbon (%; see Nelson and Sommers, 1982), nitrogen(%), and exchangeable sodium, potassium, calcium and magnesiumcations (cmol(+)/kg; see Chapman, 1965).2.3.2. Seedbank experimentIn order to explore the potential of the soil-stored seedbank for res-

toration purposes ten duplicated soil samples were taken in each of thefour pre-defined vegetation states. As seed density is highest in the top5 cmof soil (Ferrandis et al., 2001), soil samples never exceeded a depthof 10 cm and each sample of approximately 2500 cm3 was labeled.Additionally, corresponding above ground biomass was collected foreach sample designated to burn in an attempt to imitate the fuel thatwould burn in an in situ fire.

Soil samples labeled to burn (half the samples) were placed in trays(42 20 9 cm) and placed in a temporarily constructed open-airedoven. The oven was created by creating a hollowed-out berm approxi-mately (1 0.5 m) and lining it with bricks. The corresponding vegeta-tion was placed on top of the tray and ignited. Once all vegetation hadburnt the trays were removed from the enclave and allowed to coolbefore the soil was transferred to a labeled punnet (23 16 5 cm).Control samples were concurrently transferred to punnets and labeled.All 80 punnets were placed in an irrigated greenhouse and monitoredweekly for 6 months. Seedlings, once identified, were recorded andremoved.

Once the species list had been created and seedling occurrence num-bers were recorded, species were classified in the same manner as thespecies recorded in the survey of the standing vegetation communities.2.4. Statistical analyses

2.4.1. Plant community surveysFor the plant community surveys, diversity measures were calculated

for the entire study area combined and for each of the four pre-defined

image of Fig.1

Fig. 2. Photographs of the four pre-defined vegetation states found within the study area. (A = Old Renosterveld; B = Lightly Degraded Renosterveld; C = Alien Pine Stand; D= GameCamp).

138 O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145vegetation states. Alpha, beta and gamma diversity were calculated(Whittaker, 1972).

To show the arrangement of the relevs based on species compositionan ordinationwas performed using the nonmetricmultidimensional scal-ing (NMS) technique (Kruskal, 1964, Mather, 1976). In order to reducethe high coefficient of variance in the species cover matrix, specieswhich occurred in 3 or less plots were deleted. Values were then logtransformed using the following formula: transformed value = log(x +1). A matrix of abiotic values recorded from the relevs was similarlylog transformed and the overlay was used to allow for an examinationof the correlation between these variables and the ordination axes. Anal-ysis was done using PC-ORD for Windows, v 6.0, (MjM Software, 2011).

One-way analyses of variance (ANOVA)were performed on all abioticvariables. If data were not parametric, KruskalWallis one-way analysesof variance were performed. Where data were normal a post-hoc TukeyHSD test was performed.

2.4.2. Seedbank experimentAlpha diversity and gamma diversity were calculated for each vegeta-

tion state and treatment. One way analyses of variance (ANOVA) wereperformed on the Alpha diversity scores. All statistical analysis (excludingthe ordination) was performed using STATISTICA (Statsoft, 2004).

3. Results

A total of 135 different species were encountered in the 28 relevs inthe plant community survey, of which 96% were identified to specieslevel (Table 1). The mean species richness within the 28 relevs was19.20 although the standard deviation was high, indicating highvariance between relevs. The high beta diversity value of 6.03 can beexplained by the heterogeneity between the four pre-selected sites,and given the pre-selection of these different vegetation states, is ex-pected. Within the vegetation states there were clear differences: TheOld Renosterveld site (= 28.14 (sd = 4.45); = 75) and the LightlyDegraded Renosterveld (= 24.57 (sd= 3.99; =68) both exhibitedsignificantly higher alpha diversity values at the P b 0.05 level thanthose recorded in the Alien Pine stand ( = 14.43 (sd = 5.06); =41) and the Game Camp site (= 9.86 9 (sd = 1.85); = 22). The di-versity indices did not differentiate between indigenous and alien spe-cies but there was a clear correlation between low diversity valuesand alien species dominance. In both the Alien Pine stand and theGame Camp site approximately half of the species identifiedwere aliensand accounted for over 85% of the vegetation cover. In comparison inthe Old and Lightly Degraded Renosterveld sites approximately 15% ofthe species identified were alien and they accounted for less than 5%cover.

The pattern of relev ordination in Fig. 3 was not unexpected due tothe a priori selection of four study siteswith different land-use histories;however it does vindicate the decision to treat the study sites as distinct.The ordination illustrates a continuum from the most degraded GameCamp site to the more pristine Old Renosterveld site with the AlienPine stand and the Lightly Degraded Renosterveld sites falling betweenthe aforementioned along both axes.

The vegetation of the more degraded Alien Pine stand and GameCamp sites is characterized by the dominance of alien species. TheGame Camp vegetation is dominated by Avena fatua and Brachypodiumdistachyon and characterized by the presence of Echium plantagineum,whereas the presence of P. pinea is unsurprisingly a characteristic ofthe Alien Pine stand. The alien annual grass, Brizamaxima, is a dominantspecies at both sites. The Old Renosterveld site was dominated by theshrubs Passerina corymbosa and Putterlickia pyracantha and character-ized by the presence of Olea europea subsp. africana and the indigenousgrass Themeda triandra. The Lightly Degraded Renosterveld site wasdominated by Otholobium hirtum and Chrysanthemoides monilifera.Aspalathus cordata, Athanasia crithmifolia and Lobostemon fruticosuswere also shrubs typically found within Lightly Degraded Renosterveld

image of Fig.2

Table 1Species list in alphabetic order including information (where available) on plant functional type, family, national red list status, number of occurrences in relevs and dominance (N15%cover) across study sites bracketed, and presence in seedbank study. Abbreviations used: FT = functional type; LST = large shrub/tree; MHS = mid-high shrub; LS = low shrub;ALS = annual low shrubDS = dwarf shrub; PF = perennial forb; AF = annual forb; PG = perennial graminoid;AG = annual graminoid; G = Geophyte; RB = reproductive biology;RSD = re-seeder; SPR = re-sprouter; ND = no data; LC = least concern; NT = near threatened; VU = vulnerable; EN = endangered; SA = South Africa; WC = Western Cape.OR = Old Renosterveld; LD = Lightly Degraded Renosterveld; AP = Alien Pine; GC = Game Camp.

Species name FT Family Status OR LD AP GC Seedbank

Acacia cyclops A.Cunn. ex. G.Don LS/T Fabaceae Naturalized exotic 0 2 0 0 Acacia pycnantha Benth. LS/T Fabaceae Naturalized exotic 0 0 2 0 Ageratina adenophora (Spreng.) R.M.King & H.Rob. MHS Asteraceae Naturalized exotic 1 0 0 0 Anagallis arvensis L. subsp. arvensis AF Primulaceae Naturalized exotic 6 4 1 2 Arctopus echinatus L. PF Apiaceae LC-endemic to SA 0 0 1 0 Aristea africana (L.) Hoffmanns, PF Iridaceae LC-endemic to WC 2 1 0 0 Aristea pauciflora Wolley-Dod G Iridaceae ND-endemic to WC 1 0 0 0 Aspalathus astroites L. MHS Fabaceae LC-endemic to WC 1 0 0 0 Aspalathus cephalotes Thunb. subsp. violaceae LS Fabaceae LC-endemic to WC 1 0 0 0 Aspalathus chenopoda L. subsp. chenopoda MHS Fabaceae Rare-endemic to WC 0 3 0 0 Aspalathus cordata (L.) R.Dahlgren LS Fabaceae LC-endemic to WC 0 6 (1) 2 0 Aspalathus hispida Thunm. subsp. hispida MHS Fabaceae LC-endemic to SA 5 5 2 0 Asparagus capensis L. var. capensis LS Asparagaceae LC-not endemic to SA 1 0 0 0 Asparagus rubicundus P.J.Bergius MHS Asparagaceae LC-endemic to SA 0 1 0 0 Asteraceae spp. Unknown 1 AF Asteraceae 0 0 0 1 Asteraceae spp. Unknown 2 AF Asteraceae 1 0 0 0 Athanasia crithmifolia (L.) L. subsp. crithmifolia MHS Asteraceae LC-endemic to WC 1 5 0 0 Avena fatua L. AG Poaceae Naturalized Exotic 0 0 2 7 (5) Baeometra uniflora (Jacq.) G.J.Lewis G Colchicaceae LC-endemic to WC 2 0 0 0 Berkheya rigida (Thumb.) Erwart, PF Asteraceae LC-endemic to SA 1 3 0 0 Brachypodium distachyon (L.)P.Beauv

AG Poaceae Naturalized Exotic 0 0 0 5 (4)

Briza maxima L. AG Poaceae Naturalized exotic 2 1 7 (5) 7 (5) Briza minor L. AG Poaceae Naturalized exotic 0 0 0 0 Species Name FT Family Status OR LD AP GC SeedbankCarduus pycnocephalus Curtis. AF Asteraceae Naturalized exotic 0 1 4 1 Centella asiatica (L.) Urb. AF Apiaceae LC-not endemic to SA 0 0 0 0 Cheilanthes contracta (Kunze) Mett. ex Kuhn PF Sinopteridaceae LC-endemic to SA 3 0 0 0 Chenopodium hybrid (C. giganteum x C. album) AF Chenopodiaceae Naturalized Exotic 0 0 1 0 Chironia baccifera L. LS Gentianaceae LC-endemic to SA 3 4 0 0 Chrysanthemoides monilifera (L.) subsp. Monilifera MHS Asteraceae LC-endemic to SA 6 (2) 7 (3) 3 0 Chrysocoma coma-aurea LS Asteraceae Common 0 4 0 0 Cliffortia polygonifolia L. var. trifoliata Harv. MHS Rosaceae LC-endemic to WC 4 (1) 1 0 0 Clutia pulchella L. var. pulchella MHS Euphorbiaceae LC-not endemic to SA 0 1 0 0 Conyza bonariensis (L.) Cronquist AF Asteraceae Naturalized Exotic 0 0 0 0 Conyza scabrida DC. MHS Asteraceae LC-not endemic to SA 3 0 0 0 Cotula turbinata L. AF Asteraceae LC-endemic to SA 0 0 0 0 Cymbopogon plurinodis (Stapf.) PG Poaceae LC-not endemic to SA 3 1 0 0 Cynosurus echinatus L. AG Poaceae Naturalized exotic 0 0 0 0 L

Cyphia phyteuma (L.) Willd. var. phyteuma PF Lobeliaceae LC-endemic to WC 0 0 1 0 Diascia capensis (L.) Britten AF Scrophulariaceae LC-endemic to WC 0 0 0 0 Diascia diffusa Benth. AF Scrophulariaceae LC-endemic to WC 1 0 0 0 Echium plantagineum L. ALS Boraginaceae Naturalized exotic 1 0 0 7 Ehrharta calycina Sm. PG Poaceae LC-not endemic to SA 0 0 1 0 Elytropappus rhinocerotis L.f. MHS Asteraceae LC-not endemic to SA 4 7 2 0 Erodium spp. Unknown AF Geraniaceae Naturalized exotic 1 0 0 0 Euclea racemosa Murray subsp. racemosa LS/T Ebenaceae LC-endemic to SA 1 0 0 0 Euphorbia helioscopia L. AF Euphorbiaceae Naturalized exotic 0 0 0 1 Euphorbia peplus L. AF Euphorbiaceae Naturalized exotic 0 0 3 5 Euphorbia terracina L. PF Euphorbiaceae Naturalized exotic 0 2 0 0 Euryops abrotanifolius (L.) DC. LS Asteraceae LC-endemic to SA 0 0 2 0 Felicia fruticosa (L.) G.Nicholson subsp. fruticosa LS Asteraceae LC-endemic to WC 0 1 0 0 Ficinia filiformis (Lam.) Schrad. PG Cyperaceae LC-endemic to SA 4 5 0 0 Ficinia spp. Unknown PG Cyperaceae 0 0 2 0 Geranium molle L. AF Geraniaceae Naturalized exotic 1 0 0 1 Species Name FT Family Status OR LD AP GC SeedbankGeranium spp. Unknown DS Geraniaceae - 0 0 0 1 Gnidia laxa (L.f.) Gilg LS Thymelaeaceae LC-endemic to WC 0 0 0 1 Gnidia spp. unknown LS Thymelaeaceae 1 0 0 0 Helichrysum cymosum (L.) D.Don subsp. cymosum LS Asteraceae LC-endemic to SA 5 2 0 0 Helichrysum patulum (L.) D.Don LS Asteraceae LC-endemic to WC 7 7 0 0 Hermannia alnifolia L. LS Malvaceae LC-endemic to SA 0 1 0 0 Hermannia althaeifolia L. LS Malvaceae LC-endemic to SA 3 5 (1) 0 0 Hyparrhenia hirta (L.) Stapf PG Poaceae LC-not endemic to SA 7 (1) 5 (1) 0 0 Hypericum canariense L. LS/T Hypericaceae Naturalized exotic 3 0 0 0 Indigofera psoraloides (L.) L. DS Fabaceae VU-endemic to WC 2 1 0 0 Indigofera spp. Unknown PF Fabaceae 6 0 0 0 Ischyrolepis capensis PG Restionaceae LC-endemic to SA 0 1 0 0 Isolepis levynsiana Muasya AG Cyperaceae LC-endemic to SA 0 0 0 0 Isolepis marginata (Thunb.) A.Dietr AG Cyperaceae LC-not endemic to SA 0 0 0 0 Lactuca serriola L. AF Asteraceae Naturalized exotic 0 0 0 0

(continued on next page)

139O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145

Table 1 (continued)

Species name FT Family Status OR LD AP GC Seedbank

Lagurus ovatus L. AG Poaceae Naturalized exotic 1 1 0 0 Leucadendron argenteum (L.) R.Br. LS/T Proteaceae EN-endemic to WC 0 1 0 0 Lobelia anceps L.f. PF Lobeliaceae LC-not endemic to SA 0 0 0 0 Lobostemon fruticosus (L.) H.Beuk LS Boraginaceae LC-endemic to WC 0 4 (1) 0 0 Lolium perenne L. AG Poaceae Naturalized exotic 0 0 0 0 Lotononis cf. prostrata LS Fabaceae LC-not endemic to SA 0 0 0 0 Lotus angustissimus AF Fabaceae Naturalized exotic 0 0 0 0 Lythrum hyssopifolia L. AF Lythraceae Naturalized exotic 5 0 0 0 Melianthus major L. MHS Melianthaceae LC-endemic to SA 2 0 0 0 Metalasia densa (Lam.) P.O.Karis LS/T Asteraceae LC-not endemic to SA 1 3 0 0 Moraea flaccida (Sweet) Steud G Iridaceae LC-endemic to WC 4 0 0 0 Moraea spp. Unknown G Iridaceae 0 0 3 2 Moraea tricuspidata (L.f.) G.J.Lewis G Iridaceae LC-endemic to SA 1 0 0 0 Muraltia heisteria (L.) DC. MHS Polygalaceae LC-endemic to SA 0 2 1 0 Myosotis arvensis (L.) Hill AF Boraginaceae Naturalized exotic 0 2 0 0 Myrsine africana L. MHS Myrsinaceae LC-not endemic to SA 0 1 0 0 Notobubon galbanum (L.) Magee MHS Apiaceae LC-endemic to WC 0 1 0 0 Olea europea L. subsp. africana (Mill.) P.S.Green LS/T Oleaceae LC-not endemic to SA 5 0 0 0 Ornithogalum thyrsoides Jacq. G Hyacinthaceae LC-endemic to SA 0 0 0 3 Species Name FT Family Status OR LD AP GC SeedbankOtholobium hirtum (L.) C.H.Stirt. MHS Fabaceae LC-endemic to WC 0 6 (3) 1 0 Otholobium virgatum (Burm.f.) C.H.Stirt. DS Fabaceae LC-endemic to SA 0 0 0 1 Oxalis caprina L. PF Oxalidaceae LC-endemic to WC 3 1 1 4 Oxalis compressa L.f. var. compressa PF Oxalidaceae LC-endemic to WC 3 2 6 7 Oxalis corniculata L. AF Oxalidaceae Naturalized exotic 0 0 0 0 Oxalis hirta L. var. hirta PF Oxalidaceae LC-endemic to WC 0 2 0 0 Oxalis obtusa Jacq. PF Oxalidaceae LC-endemic to SA 2 0 0 0 Oxalis pes-caprae L. var. pes-caprae PF Oxalidaceae LC-not endemic to SA 1 0 0 0 Oxalis purpurea L. PF Oxalidaceae LC-endemic to SA 1 0 0 0 Oxalis tomentosa L.f. PF Oxalidaceae LC-endemic to WC 0 0 1 3 Passerina corymbosa Eckl. ex LS/T Thymelaeaceae LC-endemic to SA 7 (2) 4 0 0 Pelargonium capitatum (L.) L'Her. DS Geraniaceae LC-not endemic to SA 3 0 0 0 Pelargonium cucullatum (L.) L'Her. subsp. cucullatum MHS Geraniaceae LC-endemic to WC 1 2 0 0 Pelargonium elongatum (Cav) Salisb. DS Geraniaceae LC-endemic to SA 0 1 0 0 Pelargonium myrrhifolium (L.) L'Her. var. myrrhifolium DS Geraniaceae LC-endemic to SA 1 1 0 0 Pennisetum clandestinum Hochst. Ex Chiov. AG Poaceae Naturalized exotic 0 0 0 0 Pennisetum macrourum Trin. PG Poaceae LC-not endemic to SA 0 1 0 0 Pennisetum setaceum (Forssk.) Chiov. PG Poaceae Naturalized exotic 0 1 0 0 Pentaschistis curvifolia (Schrad.) Stapf PG Poaceae LC-endemic to SA 1 2 0 0 Phylica pubescens Aiton var. pubescens MHS Rhamnaceae LC-endemic to WC 2 1 0 0 Phytolacca octandra L. PF Phytolaccaceae Naturalized exotic 0 0 1 0 Picris echioides L. AF Asteraceae Naturalized exotic 5 2 7 0 Pinus pinea L. LS/T Pinaceae Naturalized exotic 1 3 7 (1) 0 Plantago lanceolata L. PF Plantaginaceae LC-not endemic to SA 4 2 0 0 Pseudognaphalium luteoalbum (L.) Hilliard & B.L.Burtt AF Asteraceae ND 0 0 1 0 Pseudognaphalium undulatum (L.) Hilliard&B.L.Burtt AF Asteraceae Naturalized exotic 0 0 0 0 Psoralea asarina (P.J.Bergius) T.M.Salter DS Fabaceae NT-endemic to WC 2 0 0 0 Psoralea pinnata L. var pinnata LS/T Fabaceae LC-endemic to WC 2 0 0 0 Pteridium aquilinum (L.) Kuhn subsp. PF Dennstaedtiaceae LC-not endemic to SA 1 (1) 0 0 0 Species Name FT Family Status OR LD AP GC SeedbankPteris dentata Forssk. PF Pteridaceae LC-not endemic to SA 0 0 0 0 Putterlickia pyracantha (L.) Szyszyl. LS/T Celastraceae LC-endemic to SA 5 (4) 1 0 0 Salvia africana-caerulea L. MHS Lamiaceae LC-endemic to SA 0 1 0 0 Searsia angustifolia (L.) LS/T Anacardiaceae LC-endemic to SA 1 0 0 0 Searsia lucida (L.) F.A.Barkley LS/T Anacardiaceae LC-not endemic to SA 0 4 (1) 0 0 Searsia tomentosa (L.) F.A.Barkley LS/T Anacardiaceae LC-not endemic to SA 1 1 0 0 Selago corymbosa L. LS Scrophulariaceae LC-endemic to SA 4 3 1 0 Senecio pterophorus DC. MHS Asteraceae LC-not endemic to SA 5 1 7 1 Senecio pubigerus L. LS Asteraceae LC-endemic to WC 1 3 6 4 Senecio rosmarinifolius L.f. LS Asteraceae LC-endemic to SA 1 1 0 2 Solanum marginatum L.f. MHS Solanaceae Naturalized exotic 0 1 0 0 Solanum nigrum L. AF Solanaceae Naturalized exotic 0 0 1 0 Sonchus dregeanus DC. PF Asteraceae LC-not endemic to SA 0 0 1 0 Sonchus oleraceus L. AF Asteraceae Naturalized exotic 0 0 2 0 Stoebe cinerea (L.) Thunb, MHS Asteraceae LC-endemic to SA 3 0 0 0 Taraxacum officinale Weber PF Asterceae Naturalized exotic 1 4 7 0 Themeda triandra Forssk. PG Poaceae LC-not endemic to SA 5 0 0 0 Torilis arvensis (Huds.) Link AF Apiaceae Naturalized exotic 0 4 2 0 Tribolium uniolae (L.f.) Renvoize PG Poaceae LC-endemic to SA 0 0 0 2 Trifolium angustifolium L. var. angustifolium AG Fabaceae Naturalized exotic 0 0 0 0 Trifolium dubium Sibth. AG Fabaceae Naturalized exotic 0 0 0 0 Trifolium glomeratum L. AG Fabaceae Naturalized exotic 0 0 0 0 Unknown annual forb 1 AF 0 2 3 0 Unknown annual forb 2 AF 2 0 0 0 Unknown annual forb 3 AF 0 4 0 0 Unknown annual forb 4 AF 0 1 1 0 Unknown Anthospermum spp. MHS Rubiaceae LC-endemic to SA 0 0 0 0

140 O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145

Table 1 (continued)

Species name FT Family Status OR LD AP GC Seedbank

Unknown Cyperaceae spp. AF Cyperaceae 0 2 0 0 Unknown Crassulaceae spp. LS Crassulaceae 0 0 0 0 Unknown Geophyte spp. 1 G 0 0 1 0 Unknown Geophyte spp. 2 G 1 0 0 0 Unknown Geophyte spp. 3 G 0 0 2 0 Unknown Geophyte spp. 4 G 4 0 0 0 Unknown Geophyte spp. 5 G 3 0 0 0 Species Name FT Family Status OR LD AP GC SeedbankUnknown Poaceae spp. 1 AG Poaceae 0 1 0 0 Unknown Poaceae spp. 2 AG Poaceae 0 0 1 (1) 0 Ursinia anthemoides (L.) Poir. subsp. anthemoides AF Asteraceae LC-endemic to SA 0 0 1 0 Withania somnifera (L.) Dunal LS Solanaceae LC-not endemic to SA 0 2 0 0 Zantedeschia aethiopica (L.) Spreng. G Araceae LC-not endemic to SA 7 1 0 2

141O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145relevs. Shrubs such as Aspalathus hispida, Elytropappus rhinocerotis andHelichrysum patulum; the sedge Ficinia filiformis; and the indigenousgrass Hyparrhenia hirta were species commonly found in the Old andLightly Degraded Renosterveld relevs.

The soil chemical properties from the four vegetation states showedfew significant differences to each other. The abiotic overlay on the or-dination show that an increase in exchangeable sodium cations (1.29(sd = 0.40) cmol/kg) is strongly correlated toward the Alien Pinerelevs; conversely soil resistance is strongly correlated toward OldRenosterveld while soil nitrogen levels are correlated toward LightlyDegraded Renosterveld and Old Renosterveld. Other abiotic factorsworthy ofmention are the sandy soil type found in theOld Renosterveldsite compared to the loamy soil ascribed to the other three sites andthe correlation of slope and elevation toward the Lightly DegradedRenosterveld relevs.

A total of 54 species emerged from the seedbank, 46 of which wereidentified to species level (Table 1). Twenty-seven of the identified spe-cies were indigenous and less than half the species recorded in theseedbank were identified during the plant community survey. AnFig. 3.NMS Ordination of 28 relevs in species space. The horizontal axis represents 31% of the vr2 0.3 with one ormore axis is given. Vectors are sized in proportion to their correlation co-efAP = Alien Pine Stand; GC = Game Camp.attempt to ordinate the seedbank species data proved unsuccessful asthere was no pattern evident. There were no significant differences be-tween diversity indices of the seedbanks of the different study sites ortreatments (Table 2). Neither were there any overarching patternswhen comparing treatments. In terms of alien dominance for all sitesand treatments alien seedlings accounted for more than 50% of thetotal seedlings that emerged, with the exception of the burnt OldRenosterveld treatment where alien species accounted for 45%. Onceagain there was no general pattern of the effect of the treatment onalien emergence and burning appeared to increase alien seedling emer-gence in the Lightly Degraded Renosterveld site and Alien Pine standbut decrease emergence in the Old Renosterveld and Game Campsites. In terms of actual seedling emergence a significantly greater num-ber of seedlings emerged from the Old Renosterveld site with amean of95 for the unburnt treatment and 60 for the burnt treatment. No othersite or treatment recorded a mean above 35 individual seedlings.

Fig. 4 illustrates the clear differences between the plant functionaltypes of the standing crop of the Old and Lightly Degraded Renosterveldsites, and the Alien Pine stand and Game Camp sites. The former areariance; the vertical axis represents an additional 42%. An overlay of abiotic variables withficient. Abbreviations used: OR=Old Renosterveld; LD= Lightly Degraded Renosterveld;

image of Fig.3

Table 2Mean alpha diversities (SD) and gamma diversities across study sites and by treatmentfor the seedbank. OR = Old Renosterveld; LD = Lightly Degraded Renosterveld; AP =Alien Pine Stand; GC = Game Camp.

Alpha diversity (SD) Gamma diversity

Statistic 2.74 N/ASignificance level P b 0.05 N/AOR unburnt 6.30 (1.16) 16OR burnt 8.30 (2.16) 23LD unburnt 8.00 (2.36) 32LD Burnt 5.40 (2.72) 20AP unburnt 5.10 (1.73) 18AP burnt 5.10 (1.45) 15GC unburnt 6.80 (3.00) 23GC burnt 6.40 (2.17) 21

142 O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145dominated by shrubs and have a representation of perennial graminoids,whereas the latter are dominated by annual graminoids. There are notice-able differences between the Old Renosterveld and Lightly DegradedRenosterveld site with a larger proportion of large shrub/trees re-corded in the Old Renosterveld site compared with the Lightly DegradedRenosterveld site which recorded a higher percentage of low shrubs. Therepresentation of large shrubs/trees in the Alien Pine stand was solelydown to the presence of P. pinea.

In comparison, annual forbs are farmore prominent in the seedbank.An exception to this was the Old Renosterveld site which had a highpercentage of large shrubs/trees and annual graminoids. This is attribut-able to two species, the large alien shrub Hypericum canariense and anunidentifiable species of the Cyperaceae family. There were no statisti-cally significant over-arching patterns of the effect of burning on plantFig. 4. Plant functional types and growth forms for the standing crop from the plant communitymarked b and u respectively.functional type although burning did appear to lower the proportionof annual graminoids and increase annual forbs.

4. Discussion

The results of the study provide insights to a vegetation type that ispoorly described and understood, adding to the scant knowledge ofRenosterveld dynamics in general as well as providing recommendationsfor future restoration work in the study area.

4.1. Plant community survey

The plant community survey and subsequent ordination provedwhat the initial assessment of the study area suggested in that, althoughnominally conserved as Peninsula Shale Renosterveld, the contempo-rary vegetation is a complex mosaic of indigenous and alien species. In-deed, as a result of their land-use histories, the Alien Pine stand andGame Camp sites could be described as of little conservation value andrestoration is most certainly required here to insure their conservationvalue. The Old and Lightly Degraded Renosterveld sites, while dominat-ed by indigenous species, recorded consistently lower diversity indiceswhen compared to data from similar sized relevs from South CoastRenosterveld Communities (Cowling, 1983). Indeed the gammadiversi-ty of the study area, 135 species recorded, is dwarfed by studies done inthe Renosterveld of Tygerberg (Wood and Low, 1993) and Signal Hill(Joubert, 1991) which recorded 627 and 460 species respectively; how-ever both localities encompassed a far larger area and sampling intensi-ty was greater. The relative low diversity indices may be a result of thegradual attrition in species due to the various impacts the land-use his-tory of the area and fragmentation would have had on the vegetation.surveys and the associated seedbanks. Burnt and unburnt treatments for the seedbanks are

image of Fig.4

143O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145For example at various times too frequent or conversely, the prolongedprevention of fires, may have eliminated re-seeding species from thecommunity (Cowling et al., 1997), although it has been argued thatthe reduction in re-seeding species does not necessarily reduce diversi-ty (Schwilk et al., 1997). Additionally, the effects of historical pine affor-estation and the subsequent clearingmay have reduced diversity as haspreviously been reported in Mediterranean-type ecosystems (Andresand Ojeda, 2002; Simberloff et al., 2010). Alternatively, ecosystem engi-neeringmay have contributed to the relative dearth in biodiversity: fol-lowing a fire in 1991, a heavy rain resulted in substantial mudslides andconsequently portions of the northern aspect of Devil's Peakwere delib-erately seeded with C. monilifera in a bid to stabilize the slopes (Mollet al., 1991). C. monilifera is an indigenous, early successional shrubwith weedy characteristics and is considered an invasive species inAustralia (Vranjic et al., 2000). The species' prevalence in the studyarea, and the unknown status as to which sub-species this might bewith possible varying implications, may have served to reduce diversityand it should be monitored closely to prevent it from dominating thelandscape should factors, such as increased fire frequency and distur-bance, favor its spread.

The extended absence of fire in the Old Renosterveld is another po-tential cause of concern. Renosterveld is a fire-driven system (Bondet al., 2004) with a hypothesized fire return period of between 3 and12 years (Boucher, 1983) and the dominance of larger shrubs andtrees, such as typical thicket species like P. pyracantha (Mucina andRutherford, 2006), suggests the site is well overdue for a burn and isshifting toward a thicket state. This suggested shift in state is supportedby Renosterveld vegetation models which hypothesize that in theabsence of fire Renosterveld will shift toward Subtropical thicket(Rebelo, 1995) and result in a reduction in understory diversity. To im-prove diversity and maintain the Renosterveld integrity it is recom-mended that a controlled burn is implemented in the near future.

The vegetation of the more degraded Alien Pine stand and GameCamp sites is characterized by the dominance of alien grasses. Thistrait has been reported in Renosterveld communities which have beenpreviously disturbed by overgrazing and ploughing (Boucher, 1983;Wood and Low, 1993). While the Alien Pine stand is currently beingcleared of pine trees, without management intervention the site willbe further colonized by alien grasses from the adjacent Game Campwhich has been shown to out-compete indigenous species (Musilet al., 2005).

Despite the evident differences in vegetation between the sites, thesoil chemistry was generally characterized by remarkable similarity;however significantly elevated levels of exchangeable sodium cationswere recorded in the Alien Pine stand site. The salinization of soilsunder pine plantations is a common occurrence (Berthrong et al.,2009) and is likely caused by the effect of afforestation on hydrology.Sodium is not essential for plant biochemistry; therefore plants excludeit while taking up water and other cations (Marschner, 1995). The in-creased water uptake of pines compared to indigenous species resultsin the salinization of the soil (Richardson et al., 1994). The high sodiumlevels in the Alien Pine sitemay require remediation if future restorationefforts are to be successful although, due to the high annual rainfall, thesalts may be naturally washed out the system once the pines are re-moved. If remediation is necessary, sodic soils can be treated throughthe utilization of gypsum (Calcium Sulphate) or limestone (CalciumCarbonate); however biological mediatedmeans (e.g. the buildup of or-ganic Carbon levels) are preferable since chemical additionswill changethe soil further. Heeleman et al. (2013) recorded decreased pH levels ina pine plantation in Swartland Shale Renosterveld compared to adjacentnatural veld but this was not the case here. Similarly, elevated Phos-phorous levels have been recorded in old fields in Swartland ShaleRenosterveld (Heeleman et al., 2013) and Mountain Renosterveld(O'Farrell and Collard, 2003); however this trend was absent from ourresults. The relatively uniform soil across the site excludes anylandscape-level soil remediation, and current soil conditions shouldsupport a restored Peninsula Shale Renosterveld community, pointingtoward a good starting point for any restoration initiative.

4.2. Seedbank survey

It would have been beneficial for restoration purposes if theseedbank of the degraded Alien Pine stand and Game Camp sites werefound to have an ample store of indigenous species; however this wasnot the case. Less than half the species recorded in the seedbank inthis study were identified during the plant community survey conduct-ed on the above-ground communities and the seedbank of all four siteswithin the study area were characterized by relatively low numbers(except for the Old Renosterveld site) and a dominance of alien speciesand are thus unlikely to be of any benefit to future restoration initiatives.Indeed the high volumes of competitive alien species suggest that anyecological burn in this vegetation type must be augmented by somemeans of alien clearing.

The higher seedling count found in the Old Renosterveld is largely asa result of the presence ofH. canariense, an alien invasive shrub native tothe Canary Islands. A single plant can produce hundreds of flowers, andindividual fruits can yield thousands of minuscule seeds (Dlugosch andParker, 2008), and although it was found at low densities during theplant community survey its prevalence in the seedbank of the OldRenosterveld is a cause for concern. The burning treatment significantlyreduced the emergence of the species; thus a controlled burn to redirectthe community trajectory from thicket back toward Peninsula ShaleRenosterveld may be a useful tool in reducing its dominance in theseedbank. In Australia, an infestation of the species was controlled byapplying cut-stump herbicide and basal bark treatments on the stand-ing crop and treating any re-growth prior to plants flowering and set-ting seed (Hansford and Iaconis, 2004); however there are no reportsin the literature of fire being used as a control mechanism.

Hardly any work has been carried out on Renosterveld seedbanksbut the vegetation reportedly has a higher proportion of re-sproutersin relation to re-seeders (Kemper et al., 1999). Itmay be that the subter-ranean seedbank is not as essential in post-fire recruitment as has beenobserved in Fynbos which has a high proportion of species which re-seed (Cowling et al., 1997). Conversely, work carried out in SwartlandShale Renosterveld (Heeleman et al., 2013) recorded seedbank densitiessignificantly higher than seedbank densities recorded for mountainFynbos (Holmes and Cowling, 1997) and sand plain Fynbos vegetation(Holmes, 2002) suggesting that seedbanks are an important aspect ofRenosterveld ecology. With such a dearth of studies available, it isclear that further work needs to be carried out before any clear conclu-sions can be drawn on the importance of the seedbank to Renostervelddynamics. This study suggests that for Peninsula Shale Renosterveld, inthis particular site which is degraded, the seedbank cannot be relied onfor restoration and any restoration effort must ameliorate the seedcomponent.

It is hard to draw any rigid conclusions on the role of fire onRenosterveld seedbanks due to the degraded nature of the studyarea. It may be that experimental error hid any effects of fire onseed germination as the off-site burning may not have correctly im-itated the effects of a natural wild fire. In particular, the isolated na-ture of the burn experiment may have resulted in a cooler fire. Fireintensity is a known factor in seed germination in Fynbos (Cowlinget al., 1997) and heat-stimulated germination of certain speciescan fail if the fire is too cool (Bond, 1997). Heeleman et al. (2013)treated seeds from Renosterveld with a smoke primer but, as wasthe case in this study, did not record significant overall differencesbetween treated and untreated samples. The paper acknowledgesthat any effects may have been masked by faults in the experimentaldesign; however they also speculate that Renosterveld may be moreprone to other disturbances than fire such as grazing. AlthoughRenosterveld did previously support large herds of game (Krugand Krug, 2007), its grassy nature would have promoted a short

144 O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145fire interval and it is unlikely that plants are adapted tofire aswell as graz-ingpressure. Indeed awide variety of plant species occurring infire-proneMediterranean-type vegetation such as Fynbos, Californian Chaparral(Keeley and Bond, 1997) and the vegetation in the Mediterranean Basinappear to use smoke and heat as important germination cues (Moreiraet al., 2010).

The fact that the seedbank, regardless of treatment or site status, wasdominated by alien plant species suggests that it has little value to fu-ture restoration efforts. This is contrary to findings of a similar studyin Swartland Shale Renosterveld which found that the seedbank undera pine clear-cut vegetationwas similar to that found under relative pris-tine Renosterveld (ca 70% of seedlings identified as indigenous species;(Heeleman et al., 2013)). In the adjacent Fynbos, pine clearing has led tothe recovery of a relatively diverse indigenous community (Holmeset al., 2000; Peterson et al., 2007) and pine plantations in general havebeen described as having good restoration potential due to the preser-vation of an indigenous seedbank (Kiefer and Poschlod, 1996). Never-theless there does not appear to be a viable seedbank at the study siteand, although seed dispersal from adjacent sites of indigenous vegeta-tion is an important factor in restoration (Bisteau et al., 2005), it maybe necessary to actively re-seed the Alien Pine stand site once it hasbeen completely cleared. It is also possible that these pine stands wereplanted on old lands which could account for the lower seedbank andsimilarities to the Game Camp. This is not something however thatcould be ascertained through historical records.

The poor state of the seedbank of the Game Camp site is less surpris-ing. The site was likely ploughed by the early farmers (Rebelo et al.,2011) who set up homesteads on the lower slopes and in the latterparts of the 19th century the Game Camp was cordoned off and usedas a grazing paddock. Ploughing is known to be detrimental toseedbanks (Krug and Krug, 2007) and the presence of alien grasses isknown to out-compete and restrict the establishment and growth of in-digenous species in Renosterveld depleting the seedbank over time(Musil et al., 2005).

It is clear that the current study area is a complex ecological entity as aresult of its diverse land-use history. The plant community survey identi-fied the degraded nature of the Alien Pine stand and Game Camp site andalso suggests that due to a lack of fire the Old Renosterveld site is ostensi-bly shifting toward a thicket state. Aside frommarginally elevated sodiumlevels in the Alien Pine stand, the soil chemistry of the more degradedAlien Pine and Game Camp sites was comparable to the more pristineOld and Lightly Degraded Renosterveld; thus it is unlikely that soil qualitywill restrict future Restoration endeavors. The seedbank experimentshowed no over-arching patterns between the seedbanks of the morepristine Old and Lightly Degraded Renosterveld sites and the degradedAlien Pine stand and Game Camp sites with the seedbank dominated byalien ephemeral species in almost all instances. Active restoration throughcontinued treatment and clearing of alien grass and trees followed by re-seeding programs would be necessary to return the Alien Pine stand andGame Camp sites to Peninsula Shale Renosterveld, while monitoring ofthe fire frequency of the Old and Lightly Degraded Renosterveld is themost important management tool in maintaining and promoting diversi-ty. The data provided by this study presents insights of value to futurerestoration efforts in addition to bolstering the meager knowledge onRenosterveld seedbanks.Acknowledgments

Dr. Trinder-Smith at the BolusHerbarium is thanked for assistance inspecies identification. SANParks, more specifically the staff of the north-ern section of TableMountain National Park, are thanked for their coop-eration and assistance. The financial assistance of the National ResearchFoundation and Table Mountain Fund Grant 1694 toward this researchis hereby acknowledged. Two anonymous reviewers are thanked fortheir comments on the initial manuscript.References

Anderson, P.M.L., O' Farrell, P.J., 2012. An ecological view of the history of the city of CapeTown. Ecology and Society 17, 2839.

Andres, C., Ojeda, F., 2002. Effects of afforestation with pines on woody plant diversity ofMediterranean heathlands in southern Spain. Biodiversity and Conservation 11,15111520.

Bakker, J.P., Poschlod, P., Strykstra, R.J., Bekker, R.M., Thompson, K., 1996. Seed banks andseed dispersal: important topics in restoration ecology. Acta Botanica Neerlandica 45,461490.

Berthrong, S.T., Jobbagy, E.G., Jackson, R.B., 2009. A global meta-analysis of soil exchange-able cations, pH, carbon, and nitrogen with afforestation. Ecological Applications 19,22282241.

Bisteau, E., Mahy, G., Prtel, M., 2005. Vegetation and seed bank in a calcareous grasslandrestored from a Pinus forest. Applied Vegetation Science 8, 167174.

Bond, W.J., 1997. Fire. In: Cowling, R.M., Richardson, D.M., Pierce, S.M. (Eds.), The Vegeta-tion of Southern Africa. Cambridge University Press, Cambridge, pp. 421446.

Bond, W.J., Geldenhuys, C.J., Everson, T.M., Calvin, M.F., 2004. Fire ecology: characteristicsof some important biomes of sub-saharan Africa. In: Goldammer, J.G., De Ronde, C.(Eds.), Wildland Fire Management Handbook for Sub-Saharan Africa. Global FireMonitoring Center, ComPress, pp. 1126.

Boucher, C., 1983. Floristic and structural features of the coastal foreland vegetation of theBerg River, Western Cape Province. Bothalia 14, 669674.

Bray, R.H., Kurtz, L.T., 1945. Determination of total, organic and available forms of phos-phorus in soils. Soil Science 59, 3945.

Brown, N.A.C., 1993. Promotion of germination of Fynbos seeds by plant-derived smoke.New Phytologist 123, 575583.

Chapman, H.D., 1965. Total exchangeable bases. In: Black, C.A., Evans, D.D., White, J.L.,Ensminger, L.E., Clark, F.E. (Eds.), Methods of Soil Analysis, Agronomy no. 9: chemicaland microbiological properties. American Society of Agronomy, Wisconsin, pp.902904.

Chittenden, Nicks, de Villiers, 2002. Groote Schuur Estate Conservation and DevelopmentFramework. Report (ref: 01799) for SANParks.

Clewell, A., Rieger, J.P., 1997. What practitioners need from restoration ecologists. Resto-ration Ecology 5, 350354.

Coetzee, K., 2001. Cape Peninsula National Park-Groote Schuur Estate: guidelines forwildlife introduction. Conservation Management Services, George, South Africa.

Combrink, P., Gasson, B., Oberholzer, B., Todeschini, F., Visser, D., 1992. Groote SchuurEstate Landscape Management Plan. Department of Public Works.

Cowling, R.M., 1983. Diversity relations in Cape shrublands and other vegetation in the SECape, South Africa. Vegetatio 54, 103127.

Cowling, R.M., Esler, K.J., Midgley, G.F., Honig, M.A., 1994. Plant functional diversity,species diversity and climate in arid and semi-arid southern Africa. Journal ofArid Environments 27, 141158.

Cowling, R.M., Richardson, D.M., Mustart, P.J., 1997. Fynbos. In: Cowling, R.M., Richardson,D.M., Pierce, S.M. (Eds.), Vegetation of Southern Africa. Cambridge University Press,Cambridge, pp. 99130.

Crookes, D.J., Blignaut, J.N., de Wit, M.P., Esler, K.J., Le Maitre, D.C., Milton, S.J., Mitchell, S.A.,Cloete, J., de Abreu, P., Fourie, H., Gull, K., Marx, D., Mugido, W., Ndhlovu, D., Nowell,M., Pauw, M., Rebelo, A., 2013. System dynamic modelling to assess economic viabilityand risk trade-offs for ecological restoration in South Africa. Journal of EnvironmentalManagement 120, 138147.

Diaz, S., Cadibo, M., 2001. Vive le difference: plant functional diversity matters to ecosys-tem processes. Trends in Ecology and Evolution 16, 646655.

Dlugosch, K.M., Parker, I.M., 2008. Invading populations of an ornamental shrub showrapid life history evolution despite genetic bottlenecks. Ecology Letters 11, 701709.

Euston-Brown, D. 2011. Description of natural vegetation on Groote Schuur Estate, CapeTown. University of Cape Town, unpublished report.

Ferrandis, P., De Las Heras, J., Martinez-Sanchez, J.J., Heranz, J.M., 2001. Influence of a low-intensity fire on a Pinus halepensis forest seed bank and its consequences on the earlystages of plant succession. Israel Journal of Plant Sciences 49, 105114.

Forsyth, G.G., van Wilgen, B.W., 2008. The recent fire history of the Table MountainNational Park and implications for fire management. Koedoe 50, 39.

Hannah, L., Midgley, G., Andelman, S., Araujo, M., Hughes, G., Martinez-Meyer, E., Pearson,Williams, P., 2007. Protected area needs in a changing climate. Frontiers in Ecologyand the Environment 5, 131138.

Hansford, M., Iaconis, L.J., 2004. Early intervention action on Canary Island St. JohnsWort, Hypericum canariense L., in Victoria. Fourteenth Australian Weeds Confer-ence 1, 69.

Heeleman, S., Krug, C.B., Esler, K.J., Reisch, C., Poschlod, P., 2013. Soil seedbanksof remnant and degraded swartland shale Renosterveld. Applied Vegetation Sci-ence 16, 585597.

Hobbs, R.J., Harris, J.A., 2001. Restoration ecology: repairing the earth's ecosystems in thenew millennium. Restoration Ecology 9, 239246.

Hobbs, R.J., Norton, D.A., 1996. Towards a conceptual framework for restoration ecology.Restoration Ecology 4, 93110.

Holmes, P.M., 2002. Depth distribution and composition of seed-banks in alien-invadedand uninvaded Fynbos vegetation. Austral Ecology 27, 110120.

Holmes, P.M., 2005. Results of a Lucerne old-field restoration experiment at the Fynbos/Karoo interface. South African Journal of Botany 71, 326338.

Holmes, P.M., Cowling, R.M., 1997. Diversity, composition and guild structure relationshipsbetween soil-stored seed banks and mature vegetation in alien plant-invaded SouthAfrican Fynbos shrublands. Plant Ecology 133, 107122.

Holmes, P.M., Richardson, D.M., van Wilgen, B.W., Gelderbloom, C., 2000. Recovery ofSouth African Fynbos vegetation following alien woody plant clearing and fire: impli-cations for restoration. Austral Ecology 25, 631639.

http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0005http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0005http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0010http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0010http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0010http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0015http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0015http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0015http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0020http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0020http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0020http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0025http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0025http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0030http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0030http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0035http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0035http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0035http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0035http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0040http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0040http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0045http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0045http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0050http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0050http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0290http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0290http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0290http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0290http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0295http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0295http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0055http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0055http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0060http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0060http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0065http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0065http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0070http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0070http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0075http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0075http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0075http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0080http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0080http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0080http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0085http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0085http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0085http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9000http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9000http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0090http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0090http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0095http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0095http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0095http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0100http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0100http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0300http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0300http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0110http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0110http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0110http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0115http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0115http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0115http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0120http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0120http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0125http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0125http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0130http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0130http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0135http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0135http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0140http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0140http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0140http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0145http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0145http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0145

145O.S. Cowan, P.M.L. Anderson / South African Journal of Botany 95 (2014) 135145Joubert, C., 1991. History and description of contemporary vegetation, Signal Hill, CapeTownMSc dissertation University of Cape Town, Cape Town.

Kemper, J., Cowling, R.M., Richardson, D.M., 1999. Fragmentation of South AfricanRenosterveld shrublands: effects on plant community structure and conservationimplications. Biological Conservation 90, 103111.

Kiefer, S., Poschlod, P., 1996. Restoration of fallow or afforested calcareous grasslands byclear-cutting. In: Settele, J., Margules, C., Poschlod, P., Henle, K. (Eds.), Species Survivalin Fragmented Landscapes. Kluwer, Dordrecht, pp. 209218.

Krug, C.B., Krug, R.M., 2007. Restoration in old fields in Renosterveld: a case study in aMediterranean-type shrubland of South Africa. In: Cramer, V.A., Hobbs, R.J. (Eds.),Old Fields: Dynamics and Restoration of Abandoned Farmland. Island Press, Wash-ington DC, pp. 265285.

Kruskal, J.B., 1964. Nonmetricmultidimensional scaling: a numericalmethod. Psychometrika29, 115129.

Maestre, F.T., Cortina, J., Vallejo, R., 2006. Are ecosystem composition structure and func-tional status related to restoration success? A test from semi-arid Mediterraneansteppes. Restoration Ecology 14, 258266.

Marschner, H., 1995. Mineral Nutrition of Higher Plants. Academic Press, New York.Mather, P.M., 1976. Computational Methods of Multivariate Analysis in Physical Geogra-

phy. Wiley, London.McLean, E.O., 1982. Soil pH and lime requirement. In: Page, A.L. (Ed.), Methods of Soil

Analysis, part 2: chemical and microbiological properties. American Society of Agrono-my, Wisconsin, pp. 199224.

MjM Software, 2011. PC-Ord. Multivariate analysis of ecological data. Version 6, MjMSoftware, Gleneden Beach, Oregon..

Moll, E.J., Roth, R.H., Botha, M.A., 1991. Table Mountain: distribution of alien plants. Bota-ny Department, University of Cape Town, Cape Town, pp. 19761991.

Moreira, B., Tormo, J., Estrelles, E., Pausas, J.G., 2010. Disentangling the role of heat andsmoke as germination cues in Mediterranean Basin flora. Annals of Botany 105,627635.

Mucina, L., Rutherford, M.C., 2006. The Vegetation of South Africa, Lesotho and Swaziland.Strelitzia No. 19. South African National Biodiversity Institute, Pretoria..

Musil, C.F., Milton, S.J., Davis, G.W., 2005. The threat of alien invasive grasses to lowlandCape floral diversity: an empirical appraisal of the effectiveness of practical controlstrategies: research in action. A South african Journal of Science 101, 337344.

Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Kent, J., 2000. Biodiver-sity hotspots conservation priorities. Nature 403, 853858.

Nelson, D.W., Sommers, L.E., 1982. Total carbon, organic carbon and organic matter. In:Page, A.L. (Ed.), Methods of Soil Analysis, Part 2: chemical and microbiological prop-erties. American Society of Agronomy, Wisconsin, pp. 539579.

O'Farrell, P., Collard, S., 2003. Ecosystem services: an examination of approaches to inves-tigating ecosystem services in agricultural landscapes in South Africa and Australia.Proceedings of the VIIth International Rangelands Congress, Durban (See www.nbi.ac.za/consfarm/consfarfram.htm).

Peterson, N., Husted, L., Rebelo, A.G., Holmes, P.A., 2007. Fynbos wake-up call: restorationof sand Fynbos in Tokai. Veld and Flora 93, 102.

Proches, S., Cowling, R.M., Goldblatt, P., Manning, J.C., Snijman, D.A., 2006. An overview ofthe Cape geophytes. Biological Journal of the Linnean Society 87, 2743.Rebelo, A.G., 1995. Renosterveld: conservation and research. Floral Conservation Commit-tee, Botanical Society of South Africa, Kirstenbosch, Cape Town.

Rebelo, A.G., Boucher, C., Helme, N., Mucina, L., Rutherford, M.C., 2006. Fynbos biome. In:Mucina, L., Rutherford, M.C. (Eds.), The Vegetation of South Africa, Lesotho andSwaziland. South African National Biodiversity Institute, Pretoria, pp. 158159.

Rebelo, A.G., Holmes, P.M., Dorse, C., Wood, J., 2011. Impacts of urbanization in a biodiversityhotspot: conservation challenges in metropolitan Cape Town. South African Journal ofBotany 77, 2035.

Richards, L.A. (Ed.), 1954. Diagnosis and Improvement of Saline and Alkaline Soils. USDAHandbook No. 60. U.S. Government Printing Office, Washington DC.

Richardson, D.M., Williams, P.A., Hobbs, R.J., 1994. Pine invasions in the southern hemi-sphere: determinants of spread and invadability. Journal of Biogeography 21,511527.

Rouget, M., Richardson, D.M., Cowling, R.M., Lloyd, J.W., Lombard, A.T., 2003. Current pat-terns of habitat transformation and future threats to biodiversity in terrestrial ecosys-tems of the Cape Floristic Region, South Africa. Biological Conservation 112, 6385.

Schwilk, D.W., Keeley, J.E., Bond, W.J., 1997. The intermediate disturbance hypothesisdoes not explain fire and diversity pattern in Fynbos. Plant Ecology 132, 7784.

Shaughnessy, G.L., 1980. Historical ecology of alien woody plants in the vicinity of CapeTown, South AfricaPhD dissertation University of Cape Town, Cape Town.

Simberloff, D., Nunez, M.A., Ledgard, N.J., Pauchard, A., Richardson, D.M., Sarasola, M., vanWilgen, B.W., Zalba, S.M., Zenni, R.D., Bustamente, R., Pena, E., Ziller, S.R., 2010. Spreadand impact of introduced conifers in South America: lessons from other southernhemisphere regions. Austral Ecology 35, 489504.

Society for Ecological Restoration International Policy andWorking Group, 2004. The SERInternational Primer on Ecological Restorationavailable from www.ser.org.

Statsoft, I., 2004. STATISTICA (data analysis software system), Version 7 edn. (www.statsoft.com).

South African National Botanical Institute, 2012. Redlist of South African Plants. (availablefrom www.redlist.sanbi.org).

Szabo, P., 2010.Why historymatters in ecology: an interdisciplinary perspective. Environ-mental Conservation 37, 380387.

van Wilgen, B.W., Forsyth, G.G., Prins, P., 2012. The management of fire-adapted ecosys-tems in an urban setting: the case of Table Mountain National Park, South Africa.Ecology and Society 17, 8. http://dx.doi.org/10.5751/ES-04526-170108.

Vranjic, J.A., Woods, M.J., Barnard, J., 2000. Soil-mediated effects on germination and seed-ling growth of coastal wattle (Acacia sophorae) by the environmental weed, bitoubush (Chrysanthemoides monilifera ssp. rotundata). Austral Ecology 25, 445453.

Whittaker, R.H., 1972. Evolution and measurement of species diversity. Taxon 21,213251.

Wissmar, R.C., Beschta, R.L., 1998. Restoration and management of riparian ecosystems:a catchment perspective. Freshwater Biology 40, 571585.

Wood, J., Low, B., 1993. Environmental survey and management guidelines for theTygerberg and environs. Part 1: Environmental Survey. National Botanical Institute,Cape Town.

Wyant, J.G., Meganck, R.A., Ham, S.H., 1995. The need for an environmental restorationdecision framework. Ecological Engineering 5, 417420.

http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0150http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0150http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0155http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0155http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0155http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0160http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0160http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0160http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0165http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0165http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0165http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0165http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0170http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0170http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0175http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0175http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0175http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0180http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9080http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9080http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0185http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0185http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0185http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0305http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0305http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0305http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0310http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0310http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0195http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0195http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0195http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0315http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0315http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0315http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9880http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9880http://refhub.elsevier.com/S0254-6299(14)00173-2/rf9880http://refhub.elsevier.com/S0254-6299(14)00173-2/rf8800http://refhub.elsevier.com/S0254-6299(14)00173-2/rf8800http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0200http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0200http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0200http://www.nbi.ac.za/consfarm/consfarfram.htmhttp://www.nbi.ac.za/consfarm/consfarfram.htmhttp://refhub.elsevier.com/S0254-6299(14)00173-2/rf0205http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0205http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0210http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0210http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0215http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0215http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0325http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0325http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0325http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0225http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0225http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0225http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0330http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0330http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0230http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0230http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0230http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0235http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0235http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0235http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0240http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0240http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0245http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0245http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0250http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0250http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0250http://www.ser.orghttp://www.statsoft.comhttp://www.statsoft.comhttp://www.redlist.sanbi.orghttp://refhub.elsevier.com/S0254-6299(14)00173-2/rf0260http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0260http://dx.doi.org/10.5751/ES-04526-170108http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0265http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0265http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0265http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0275http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0275http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0270http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0270http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0345http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0345http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0345http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0285http://refhub.elsevier.com/S0254-6299(14)00173-2/rf0285

The Peninsula Shale Renosterveld of Devil's Peak, Western Cape: A study into the vegetation and seedbank with a view toward...1. Introduction2. Materials and methods2.1. Site description2.2. Site history2.3. Data collection2.3.1. Plant community surveys2.3.2. Seedbank experiment

2.4. Statistical analyses2.4.1. Plant community surveys2.4.2. Seedbank experiment

3. Results4. Discussion4.1. Plant community survey4.2. Seedbank survey

AcknowledgmentsReferences