An Assessment of a Proposal to Eradicate Non-Native Fish From

This article was downloaded by: [University of Stellenbosch] On: 25 January 2013, At: 06:30 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

African Journal of Aquatic Science Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/taas20 An assessment of a proposal to eradicate non-native fish from priority rivers in the Cape Floristic Region, South Africa SM Marr a , ND Impson b & D Tweddle c a Freshwater Research Unit, Zoology Department, University of Cape Town, Private Bag X3, Rondebosch, 7700, South Africa b CapeNature, Private Bag X5014, Stellenbosch, 7599, South Africa c South African Institute for Aquatic Biodiversity, Private Bag 1015, Grahamstown, 6140, South Africa Version of record first published: 29 Jun 2012.

To cite this article: SM Marr , ND Impson & D Tweddle (2012): An assessment of a proposal to eradicate non-native fish from priority rivers in the Cape Floristic Region, South Africa, African Journal of Aquatic Science, 37:2, 131-142 To link to this article: http://dx.doi.org/10.2989/16085914.2012.666654

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. African Journal of Aquatic Science 2012, 37(2): 131–142 Copyright © NISC (Pty) Ltd Printed in South Africa — All rights reserved AFRICAN JOURNAL OF AQUATIC SCIENCE ISSN 1608-5914 EISSN 1727-9364 http://dx.doi.org/10.2989/16085914.2012.666654

An assessment of a proposal to eradicate non-native fish from priority rivers in the Cape Floristic Region, South Africa

SM Marr1*, ND Impson2 and D Tweddle3

1 Freshwater Research Unit, Zoology Department, University of Cape Town, Private Bag X3, Rondebosch 7700, South Africa 2 CapeNature, Private Bag X5014, Stellenbosch 7599, South Africa 3 South African Institute for Aquatic Biodiversity, Private Bag 1015, Grahamstown 6140, South Africa * Corresponding author, e-mail: [email protected]

Non-native fish are considered the most important threat to the survival of the indigenous freshwater fishes in the Cape Floristic Region (CFR). A pilot project to evaluate the use of the piscicide rotenone to eradicate non-native fish from selected reaches in four rivers has been proposed by CapeNature, the conservation authority of the Western Cape province, South Africa. Each river has unique characteristics and challenges to achieving the eradication of non-native fish and the restoration of its indigenous fish assemblage. The proposal is described and the management methods available for reducing the impact of non-native fish on indigenous species are discussed. An environmental impact assessment (EIA) concluded that the project was justified and necessary, the choice of rivers sound, and supported the use of piscicides. The need for the project, the selection of sites and the findings of the EIA are discussed. It is expected that the project will be successful while having minimal impact on other aquatic fauna. The successful completion of the pilot project could help establish methods to eradicate non-native fish from critical biodiversity areas in South Africa. Such projects must, however, be part of a comprehensive conservation management plan to be implemented by conservation agencies in the CFR.

Keywords: conservation, eradication, piscicides, river rehabilitation

Introduction

The Cape Floristic Region (CFR), an 88 000 km2 (2011) recognise 43 genetic lineages of primary freshwater Mediterranean-climate region on the south-western tip of fishes in the CFR. Of the currently recognised 27 taxa, six Africa, is a recognised biodiversity hotspot with high levels are listed as Critically Endangered (22%), ten as Endangered of endemicity (Myers et al. 2000a, Cowling and Pressey (37%), three as Vulnerable (11%), three as Near Threatened 2003, Darwall et al. 2009). Internationally acclaimed for its (11%), two as Least Concern (7%), and three as Data plant biodiversity, the CFR is also a centre of endemism for Deficient (11%) in the 2007 IUCN assessment (Impson freshwater fauna, including aquatic invertebrates, amphib- 2007, Tweddle et al. 2009). Diadromous taxa are present ians and fishes (Skelton et al. 1995, Wishart and Day 2002, in the CFR but are not addressed in this paper because all Younge and Fowkes 2003, Darwall et al. 2009). The propor- are currently listed as Least Concern (Tweddle et al. 2009). Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 tion of endemic freshwater species is higher than that of With the imminent recognition of additional endemic taxa, plants, although species numbers are lower (Wishart and most with extremely limited distributions, the present number Day 2002). The freshwater fishes of the CFR exhibit typical of threatened taxa is considered conservative (Swartz and characteristics of old, well-established mountain faunas, Tweddle 2011). The 2007 IUCN assessment lists the major including a high degree of endemicity (89%), geographically threats to the CFR’s 24 endemic primary freshwater fish taxa restricted ranges, relative inflexibility in life-history styles, as: non-native fish (23 taxa), habitat destruction (18 taxa), and a low resilience to disturbance, especially by introduced pollution including pesticides (2 taxa), utilisation (3 taxa) and piscivorous fish (Skelton 1987). Most species are relatively genetic integrity (4 taxa) (Tweddle et al. 2009). Researchers small, <150 mm SL when adult. None are entirely piscivo- and conservation officials agree that non-native fish are the rous, although the Cape kurper Sandelia capensis (Cuvier most significant threat to the long-term survival of indigenous 1831) and some of the larger cyprinids, e.g. Clanwilliam freshwater fish assemblages in the CFR (Barnard 1943, yellowfish Labeobarbus capensis (A. Smith 1841), are partly Coke 1988, Skelton et al. 1995, Impson and Hamman 2000, piscivorous as adults (Skelton 2002). Skelton 2000, Impson et al. 2002a, 2002b, Skelton 2002, Twenty-seven taxa of primary freshwater fish are currently Cambray 2003a, 2003b, Impson 2007, Tweddle et al. 2009, recognised for the CFR (Impson 2007, Tweddle et al. 2009), Swartz and Tweddle 2011). but this number will increase following taxonomic revisions of A proposal by CapeNature, the conservation agency for the genera Pseudobarbus Smith 1841, Galaxias Cuvier 1816 the Western Cape province (South Africa), to eradicate non- and Sandelia Castelnau 1861, and Swartz and Tweddle native fish from four rivers in the CFR is described. The

African Journal of Aquatic Science is co-published by NISC (Pty) Ltd and Taylor & Francis 132 Marr, Impson and Tweddle

proposal justifies the need for the project and the site escapees into natural watercourses could have an impact selection, and reports the findings of the environmental on both aquatic vegetation and invertebrates (Pípalová impact assessment undertaken to review the project. 2006). This paper reviews available literature on the impacts of Three endemic taxa have been translocated between non-native fish on indigenous freshwater fishes in the catchments within the CFR. Cape kurper was illegally CFR and discusses the management methods available to introduced into a dam in the Twee River catchment (Olifants– reduce those impacts. Doring System) in the 1950s (Hamman et al. 1984) and subsequently invaded this catchment following the rupture Non-native fish introductions of the dam during a rainfall event (Impson et al. 2007, Marr et al. 2009). In the 1980s, the conservation authority in the The CFR has a long history of freshwater fish introductions Western Cape, the then Cape Department of Nature and (de Moor and Bruton 1988, Skelton 1990, Skelton et al. Environmental Conservation, intentionally introduced the 1995, Skelton 2000, 2002), dating back to those by Dutch Clanwilliam yellowfish above waterfall barriers in the Twee and English settlers in the late 1700s (de Moor and Bruton (Impson et al. 2007), Ratels (Impson 2010) and Boontjies 1988, Picker and Griffiths 2011). To date, 24 species of rivers (Impson and Tharme 1998) in the Olifants–Doring freshwater fish have been introduced to the inland waters of catchment as a conservation measure. Breede River redfin the CFR for recreational angling, aquaculture, or the biocon- Pseudobarbus sp. ‘burchelli Breede’ were unwisely cultured trol of aquatic weeds and mosquitoes (Table 1). Of these, by the conservation authority at Jonkershoek hatchery in seven have failed to establish self-sustaining populations outdoor ponds, from which they escaped and invaded the and one, Eurasian perch Perca fluviatilis Linnaeus 1758, is Eerste River near Stellenbosch in the 1980s. believed to have been extirpated. In addition, triploid grass In the 1980s the conservation authority declared a carp Ctenopharyngodon idella (Valenciennes 1844) have moratorium on the stocking of fish into the rivers of the CFR been introduced to farm dams for aquatic weed control. (Coke 1988, Skelton 2000). Since 1988 no new species Although these fish are sterile (a requirement of permits), have been recorded, but the secondary spread of species they are long-lived, grow to be very large (over 1 m), and previously introduced continues unabated (Impson et al.

Table 1: Summary of freshwater fish introductions into the Cape Floristic Region (after de Moor and Bruton 1988, Skelton 2001)

Year Species Reason for introduction Outcome 1700s Carp (Cyprinus carpio) Ornamental, aquaculture, angling Successful Goldfish (Carassius auratus) Ornamental Successful 1890s Brown trout (Salmo trutta) Angling Successful Rainbow trout (Oncorhynchus mykiss) Angling, aquaculture Successful Atlantic salmon (Salmo salar) Angling Failed 1910 Tench (Tinca tinca) Angling Successful 1912 Guppy (Poecilia reticulata) Biological control (mosquitoes) Failed 1915 Israeli tilapia (Oreochromis aureus) Aquaculture Successful Eurasian perch (Perca fluvialitis) Angling Successful + 1928 Largemouth bass (Micropterus salmoides) Angling Successful 1936 Mozambique mouthbrooder (Oreochromis mossambicus) Aquaculture Successful Mosquitofish (Gambusia affinis) Biological control (mosquitoes) Successful 1937 Smallmouth bass (Micropterus dolomieu) Angling Successful Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 1938 Bluegill (Lepomis macrochirus) Fodder for angling species Successful 1939 Spotted bass (Micropterus punctulatis) Angling Successful 1941 Banded tilapia (Tilapia sparrmanii) Fodder for angling species Successful 1950 Brook char (Salvelinus fontinalis) Angling Failed 1953 Smallmouth yellowfish (Labeobarbus aeneus) Angling Successful 1959 Nile tilapia (Oreochromis niloticus) Aquaculture Failed ++ Mango tilapia (Sarotherodon galilaeus) Aquaculture Failed Red-bellied tilapia (Tilapia zillii) Biological control (aquatic plants) Failed 1980s Sharptooth catfish (Clarias gariepinus) Aquaculture, angling Successful Grass carp (Ctenopharyngodon idella) Biological control (aquatic plants) Sterile triploids +++ Southern mouthbrooder (Pseudocrenilabrus philander) Fodder for angling species Successful + Eurasian perch successfully established in two systems in the CFR, but has since been extirpated from both systems. One system was treated with rotenone and drained. ++ Nile tilapia was reported as being established on the Cape Flats near Cape Town, but recently no populations of this species have been recorded. Consequently it is believed that this species did not become established in the CFR, even though climate models predict that it should be able establish in the region (Zambrano et al. 2006). The aquaculture industry is currently pushing for permission for hybrid O. mossambicus u O. niliticus to be introduced into the CFR. +++ Grass carp has been introduced into water bodies in the CFR to control aquatic vegetation. Only sterile triploid fish from certified hatcheries were permitted to be introduced. The species has established in the Orange–Vaal and Pongolo systems, even though they were expected not to be able to establish there. The species is long-lived and could survive in natural water courses if they escape from the farm dams. African Journal of Aquatic Science 2012, 37(2): 131–142 133

2002a, 2002b, Skelton 2002, Impson 2006, 2007). Of partic- indigenous and non-native fishes is particularly evident ular concern is the sharptooth catfish Clarias gariepinus where centrarchids (bass and sunfish) and salmonids are (Burchell 1822), which has been introduced into several present. Non-native fish affect the behaviour and composi- river systems by anglers, farm labourers, rural landowners, tion of indigenous fish assemblages in the CFR (Cambray and via inter-basin transfer schemes (Cambray 2003c, 2003a, 2003b, Woodford and Impson 2004, Woodford et al. Impson 2006, 2007). It is possible that additional taxa have 2005, Shelton et al. 2008) as well as affecting lower trophic been introduced into the region, but these introductions levels of the food web including aquatic invertebrates and have not yet been confirmed by researchers or conserva- algae (Lowe et al. 2008). Because of this, the remaining tion officials. indigenous fish populations can only be conserved, and It has been estimated that more than 90% of mainstem their ranges increased, through the elimination of non-native river habitat in the CFR has been invaded by non-native species from those rivers identified as conservation priori- fish and only seven small coastal catchments, together ties (Impson 2007). At the national level, critical biodiversity representing 0.8% of the CFR by area, have no recorded areas for indigenous freshwater fauna have been mapped non-native fish introductions (Marr 2012). The majority of through the National Freshwater Ecosystem Priority Areas catchments contain four or more non-native species, with the (NFEPA) project (Driver et al. 2011, Nel et al. 2011), which major river systems of the Olifants–Doring, Berg, Breede and identified priority areas for freshwater biodiversity conser- Gouritz containing 10 or more non-native species (Figure 1). vation that require, among other intervention measures, Indigenous primary freshwater fishes are absent, or rare, in non-native fish control/eradication. the majority of the reaches invaded by non-native species, and are largely restricted to reaches above barriers that have Management of non-native fish populations in the CFR prevented invasion by non-native species (Skelton 1990, Skelton et al. 1995, Skelton 2000, Impson et al. 2002b, Although the presence of predatory non-native species in Skelton 2002, Impson 2007, Tweddle et al. 2009, Swartz and the CFR poses the greatest of several threats to indigenous Tweddle 2011). fish assemblages (Impson and Hamman 2000, Impson et al. 2002b, Skelton 2002, Impson 2007, Tweddle et al. Impacts of non-native fish on indigenous fishes in the CFR 2009, Swartz and Tweddle 2011), it is acknowledged that non-native freshwater fishes such as trout, bass and carp Freshwater communities are particularly vulnerable to non- are the mainstay of South Africa’s recreational angling and native species introductions (Saunders et al. 2002, Skelton freshwater aquaculture industries (van Rensburg et al. 2002, Cambray 2003a, 2003b). The introduction of non-native 2011). A balance must therefore be maintained between fish results in impacts at genetic (gene transcription, hybridi- the conservation of indigenous taxa and the management sation), individual (behaviour, morphology, metabolic rates), of non-native fish (Chadderton 2003, van Rensburg et population (transmission of parasites/diseases, demographic al. 2011). This balance is an important aim of the regula- effects, distributional effects), community (species extirpa- tions being developed as part of the National Environmental tions, compositional changes, alterations in food webs) and Management: Biodiversity Act (NEM:BA; RSA 2004). ecosystem (biochemical cycles, energy fluxes between NEM:BA divides non-native species into three catego- ecosystems, ecological engineering) levels (Cucherousset ries: prohibited species which may not be imported into and Olden 2011). the country, exempted species which may be freely traded The negative ecological impact of non-native fish on within the country without permit, and invasive non-native indigenous fishes was first recognised in the late 1930s, species which must be controlled by means of implementing especially in the biodiversity hotspot of the Cederberg, a management plan. The latter category includes the Western Cape (Barnard 1943). Predation by non-native species most commonly targeted by anglers, such as brown Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 fish on indigenous fish assemblages was raised as a signif- trout Salmo trutta Linnaeus 1758, rainbow trout, largemouth icant concern in the CFR when the predatory effects of bass Micropterus salmoides (Lacepède 1802), smallmouth rainbow trout Oncorhynchus mykiss (Walbaum 1792) and bass and carp Cyprinus carpio Linnaeus 1758. For all other smallmouth bass Micropterus dolomieu Lacepède 1802 non-native species, a risk assessment is required prior on Pseudobarbus redfin were noticed (Hey 1926, Barnard to the issue of permits (van Rensburg et al. 2011). It is 1943, Harrison 1950, 1951a, 1951b, 1952a, 1952b, Skelton envisaged that an area approach (i.e. zoning) will be part 1983, Coke 1988, Skelton 1990, Skelton et al. 1995, of the management plans developed for economically and Skelton 2000, 2002). Despite this, non-native fish continued recreationally important non-native species. The zoning of to be stocked throughout the CFR until the 1980s (Coke areas for the use of various non-native species allows for 1988, Impson and Hamman 2000, Skelton 2000, Impson trade-offs between conservation priorities (e.g. NFEPA) et al. 2002a, Richardson et al. 2004, Tweddle et al. 2009). and recreational-economic interests (van Rensburg et al. However, the initial impact of non-native fish on indigenous 2011), and has been completed in an open participatory fish assemblages in the CFR was poorly recorded and not basis involving experts on aquatic conservation issues and studied in detail, because most impacts occurred before any representatives from key angling and aquaculture groups. great interest in conserving the indigenous taxa arose. Prevention of future introductions is the most effective There is a strong correlation between the presence of way of addressing invasion by non-native species predatory non-native fish and the absence of indigenous (Saunders et al. 2002, Britton et al. 2011a) and can be taxa, a situation analogous with that in the Colorado River, enhanced by instituting adequate decision support tools USA (Marsh and Pacey 2005). This immiscibility between and risk assessment metrics (Britton et al. 2011a, 2011b). 134 Marr, Impson and Tweddle 33° S Africa South AFRICA Krom (Eastern Cape) Berg Breede Gouritz Olifants-Doring Gamtoos Catchment A B C D E INDIAN OCEAN SOUTH AFRICA ered for the eradication of non-native fish E Groot Paradys (Gouritz) Dorps tributary of (Gouritz) D No non-native species 1–3 non-native species 4–6 non-native species 7–9 non-native species 10–14 non-native species Twee Matjies Suurvlei (Cederberg) Krom tributary of (Cederberg) tributary of Rondegat (Cederberg) 20° E C A INDIAN OCEAN B Map of the Cape Floristic Region reflecting numbers non-native fish per catchment and locations six rivers consid OCEAN

ATLANTIC

Figure 1: Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 2013 January 25 06:30 at Stellenbosch] of [University by Downloaded African Journal of Aquatic Science 2012, 37(2): 131–142 135

Capacity at relevant levels (i.e. law enforcement, taxonomic of other non-native fish species prevented. Education expertise and communication) is critical for prevention and publicity initiatives with local landowners and angling programmes to be effective. Where non-native species bodies should be established, in conjunction with simple have already become established, active management and inexpensive monitoring protocols to detect non-native needs to focus on reducing their impacts and preventing species in restored reaches, to ensure long-term success further spread (Saunders et al. 2002, Britton et al. 2011a). of conservation programmes (Clarkson et al. 2005). Eradication is more cost-effective than long-term mechan- Establishing privately owned protected areas through ical control (Bomford and Tilzey 1997, IUCN 2000), particu- stewardship initiatives, as part of South Africa’s Protected larly for recent introductions, or where the species has been Areas Act (RSA 2003), is necessary to ensure the active spatially constrained (Britton et al. 2011a). Eradication of conservation of the indigenous freshwater fishes of the CFR. non-native fishes is achieved by chemical treatments, The NFEPA project and the zoning of catchments e.g. piscicides, or by draining water-bodies (Finlayson et satisfied the first step in the above management strategy. al. 2000, Collares-Pereira and Cowx 2004, Britton et al. The next step was to begin active non-native fish eradica- 2011a). Sustained mechanical removal has been successful tion projects in priority conservation areas. In order to only at small scales (Britton et al. 2011a). Eradication achieve this, the CAPE Alien Fish Eradication Project was should, however, only be attempted where it is ecologically proposed as a pilot project to evaluate the use of piscicides feasible and has financial and political support (Huntley for the eradication of non-native fish from selected rivers 1999, IUCN 2000). Where eradication is not feasible, in the CFR. This was done to comply with section 28 of control is the next best alternative. Control programmes South Africa’s National Environmental Management Act using mechanical removal techniques (e.g. electric fishing (NEMA; RSA 1998a), which obliges conservation authori- or netting) are generally effective in suppressing popula- ties to remedy degradation of environments harmed by non- tion abundance and reducing their recruitment (Britton et al. native and/or invasive species. 2011a). Eradication and control of non-native fishes remain constrained by their lack of selectivity, and the challenges The CAPE Alien Fish Eradication Project of treating large spatial scales effectively (Britton et al. 2011a). Non-native species invasions are, however, The Cape Action Plan for People and the Environment generally irreversible (Cucherousset and Olden 2011) (CAPE) Project, a comprehensive conservation plan for the and the current technologies available for the eradication CFR (Younge and Fowkes 2003), recognised the need for of established non-native species are likely to result in the intervention to conserve the freshwater fishes of the CFR loss of some indigenous taxa from treated reaches during and established the CAPE Alien Fish Eradication Project the eradication process (Myers et al. 2000b), although most to identify priority rivers for non-native fish eradication will recolonise the treated reaches from adjacent areas (Impson 2007). Funded by the World Bank and managed (Vinson et al. 2010). by CapeNature (formerly the Cape Department of Nature While large-scale eradication of non-native fish is difficult, and Environmental Conservation), this project comprised and rarely implemented, small-scale projects to eradicate two phases (Impson 2007): non-native fish from priority reaches of small rivers to • Phase 1 — identification of priority rivers for the eradica- re-establish threatened fish species to parts of their original tion of non-native fish, plus completion of an environ- distribution range, and/or increase their distribution range, mental impact assessment (EIA) for the most suitable can be successfully completed with currently available rivers, and technologies. For example, a piscicide-based conserva- • Phase 2 — execution of the intervention, if deemed to be tion management strategy, such as the one proposed for effective and appropriate. the control of non-native fish in the lower Colorado River The pilot project aims to evaluate the eradication of Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 (Clarkson et al. 2005), could be implemented as follows: non-native fish from four streams in the CFR using the (a) Conservation authorities and other stakeholders identify piscicide rotenone and to monitor the subsequent recovery rivers to be devoted exclusively to the conservation of of the treated reaches, focusing on the recovery of threat- indigenous fishes, and those for non-native fish utilisa- ened indigenous fishes. If successful, the project will tion, e.g. aquaculture and angling. provide protocols for the implementation of active indige- (b) In rivers identified as conservation priorities where nous fish conservation projects in the CFR, and elsewhere non-native species are present, beginning from head- in South Africa. waters, temporary barriers are installed and any indigenous fish above them are caught and retained in Selection of rivers off-stream holding facilities. Criteria for the selection of rivers appropriate for the pilot (c) Non-native species above the barrier are chemically project were determined at a specialist workshop held removed, after which the indigenous fish are returned, at the South African Institute for Aquatic Biodiversity, or introduced from an appropriate stock. Grahamstown (Impson 2007). The criteria included biolog- (d) The process is then successively repeated downstream, ical, land use, social, financial and logistical considera- removing the temporary upstream barrier after each tions (Impson 2007). Field surveys of all potential rivers step and eradication of the non-native species has identified at the workshop were conducted in March and been confirmed. April 2004 to evaluate them against these criteria. A The risk of the illegal release of non-native fish into the subsequent workshop selected the following six rivers for treatment area must be minimised and new introductions the pilot project: the Rondegat, Krom and Suurvlei rivers 136 Marr, Impson and Tweddle

in the Cederberg; the Dorps and Paradys rivers in the and potential social issues regarding the use of a piscicide in Gouritz catchment; and the Krom River in the Eastern Cape the water supply of a town in a water-limited area. (Figure 1). Further field surveys, using a combination of fyke netting, seine netting, electric fishing and snorkel surveys, Rondegat River, Cederberg were conducted between August 2005 and October 2006 The Rondegat River, a tributary of the Olifants, drains a to delineate the distribution ranges of indigenous and non- moderately transformed catchment containing pristine natural native fish and to identified potential barriers that could be areas, cattle pastures and fruit orchards, before discharging used as the upper or lower barriers for the eradication of into Clanwilliam Dam. Above a small waterfall barrier 5 km non-native fish by using piscicides. above Clanwilliam Dam (Figure 2), the river holds healthy At the third specialist workshop, the list was reduced to the populations of five indigenous fish taxa: Clanwilliam redfin four rivers believed to offer the highest likelihood of success, Barbus calidus Barnard 1938, fiery redfin Pseudobarbus i.e. the Krom, Rondegat and Suurvlei rivers in the Cederberg, phlegethon (Barnard 1938), Clanwilliam rock catfish and the Krom River (Eastern Cape). Each river had different Austroglanis gilli (Barnard 1934), Cape galaxias Galaxias characteristics and conservation targets. The Dorps and zebratus Castelnau 1861 (probably an undescribed taxon, Paradys rivers, in the Gouritz catchment, were considered E Swartz, South African Institute for Aquatic Biodiversity, unsuitable for the pilot project due to the lack of suitable pers. comm.), and Clanwilliam yellowfish (Woodford et al. barriers limiting the upper range of non-native fish invasion, 2005). Below the barrier, only large Clanwilliam yellowfish poor accessibility to the river to implement the eradication, and invasive smallmouth bass are found. A water abstraction

Upstream of Abstraction Waterfall waterfall barrier weir barrier

19° E Rondegat Clanwilliam yellowfish 0 5 10 km Labeobarbus capensis

Clanwilliam Dam

EnlargedEnlarged areaarea Abstraction weir 52°15ƍ S Downstream of Clanwilliam rock catfish Austroglanis gilli Waterfall barrier waterfall barrier

Rondegat Downloaded by [University of Stellenbosch] at 06:30 25 January 2013

Clanwilliam yellowfish Fiery redfin Olifants Labeobarbus capensis Pseudobarbus phlegethon Algeria campsite

Clanwilliam redfin Barbus calidus

Smallmouth bass* Micropterus dolomieu

Cape galaxias Galaxias zebratus

Figure 2: Map of the Rondegat River showing landmarks, barriers and fish distributions (* denotes non-native species) African Journal of Aquatic Science 2012, 37(2): 131–142 137

weir, situated 4 km downstream of the barrier, marks the reportedly common in pools near the campsite before the lower boundary of the proposed intervention area. Below this introduction of trout, but now appear to have been extirpated. weir, the river is dry for most of the summer months. The weir Three non-native species have been introduced: rainbow will require reinforcement, however, to prevent re-invasion trout in 1957 (Weaver 2008), and bluegill sunfish Lepomis by bass or other species in the medium- to long-term. macrochirus Rafinesque 1819 and largemouth bass in the The river flows in a single channel through the proposed 1980s. Rainbow trout occur from Disa Pool to just above intervention area, although several furrows leading off it are the Matjies River confluence, while bluegill sunfish and used to irrigate riparian pastures. If successful, the project largemouth bass occur in farm dams and in the river itself will extend the distribution of the five indigenous taxa in the from the campsite downstream. A derelict weir just above Rondegat River by more than 20%, providing more varied the confluence with the Matjies River is the potential lower habitat with larger, deeper pools, and establishing a buffer boundary for the non-native fish eradication, but would require zone between the non-native species and the indigenous refurbishment to be an effective barrier to non-native fish. taxa. It could also allow the re-introduction of Clanwilliam Non-native fish eradication from the Krom River would sawfin Barbus serra Peters 1864 and Clanwilliam sandfish provide an opportunity to establish a sanctuary for the Labeo seeberi (Gilchrist & Thompson 1911), reported by van Critically Endangered Doring fiery redfin, an undescribed Rensburg (1966) from the lower reaches of the Rondegat but taxon restricted to two unsustainable populations in the extirpated following bass invasion. The re-establishment of Doring system, both severely threatened by bass. With these two endangered taxa would be a major conservation its upper reaches lying in a private conservancy, farming success. activities at Krom River Farm being scaled back in favour of ecotourism, and its lower reaches in the Matjies Nature Krom River, Cederberg Reserve, the Krom River has the potential to be a ‘flagship’ The Krom River is a tributary of the Matjies River in the project for fish conservation and environmental awareness, Doring catchment. A large waterfall above Disa Pool both in the CFR and in South Africa as a whole. (Figure 3) marks the upper limit of fish distribution. Below Disa Pool the river flows through a gorge of bedrock steps, Suurvlei River, Cederberg pools and chutes. The valley opens up below the gorge The Suurvlei River, a tributary of the Twee River in the and the low-gradient river, with sandy runs and pools, flows Doring River catchment, drains an area of intense citrus through a near-pristine valley before entering the highly and soft fruit farming. Two fish, Twee River redfin Barbus transformed campsite section at Krom River Farm. Below the erubescens Skelton 1974 and an undescribed taxon of farm, the river flows through the near-pristine Matjies River Galaxias (E Swartz pers. comm.), are the only indigenous Nature Reserve to its confluence with the Matjies River. One fish taxa in the Twee River, both being endemic to this small indigenous species, the Clanwilliam rock catfish, occurs from catchment. Of these, only the redfin has been recorded from the lower reaches of the gorge to just above the campsite the Suurvlei tributary in recent surveys (Impson et al. 2007, (Figure 3). Redfin, possibly Clanwilliam redfin and/or Doring Marr et al. 2009). Three non-native fish taxa, translocated fiery redfin Pseudobarbus sp. ‘phlegethon Doring’, were CFR endemics Cape kurper and Clanwilliam yellowfish, and

32°30ƍ S 19°15ƍ E Matjies

Sneeuberg Krom River Farm and camp site Proposed

Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 barrier weir Upper gorge Krom

Waterfall barrier Matjies River Disa Pool Nature Reserve Distribution range of Clanwilliam rock catfish Farm dams with bluegill and bass

Road Clanwilliam rock catfish 0 2.5 5 km River Austroglanis gilli

Figure 3: Map of the Krom River (Cederberg) showing landmarks, barriers and fish distributions 138 Marr, Impson and Tweddle

19°10ƍ E

Suurvlei

Twee River redfin Barbus erubescens

Buffelshoek

Suurvlei waterfall

Redfin range Road to Citrusdal Cape kurper* Sandelia capensis

Cape kurper range

Suikerbossie bridge

Road Suurvlei 32°40ƍ S River 32°40ƍ S

0 2 4 km

19°10ƍ E

Figure 4: Map of the Suurvlei River showing landmarks, barriers and fish distributions (* denotes non-native species)

North American bluegill sunfish, are present in the upper occur in the headwaters of the Krom River, the Krom Twee catchment (Impson et al. 2007, Marr et al. 2009). River redfin Pseudobarbus sp. “afer Krom” and a Galaxias Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 Only Cape kurper have been recorded in the proposed taxon. Galaxias were recorded above a large waterfall, intervention area (Figure 4). The upper limits of Cape kurper while redfin occurred from below the waterfall downstream are a culvert on the Buffelshoek tributary and a bedrock step to a set of weirs built in 2001 by the Working for Wetlands on the Suurvlei River. A small population of redfin persists programme (Figure 5). Redfin were also found in a tributary, above the Cape kurper distribution in the Suurvlei River downstream of the weirs, currently protected from bass (Figure 4). The lower limit of the proposed intervention is a invasion by the water abstraction off-take weir. Largemouth bedrock step at Suikerbossie Bridge (Figure 4). However, bass are present in farm dams above the redfin distribution this barrier is considered insufficient to prevent re-invasion by area and could invade the upper reaches through irriga- Cape kurper and the construction of a weir is recommended tion furrows although large floods appear to flush them from for the site. Eradication of the Cape kurper from the Suurvlei the ‘flashy’ upper reaches into the slower-flowing middle will extend available habitat for the Twee River redfin and reaches, where they are common. The eradication project will allow the re-introduction of Galaxias, but this should be proposes to remove bass from between the waterfall and regarded as only the first step in a more comprehensive the Working for Wetlands weirs and from the farm dams conservation programme for the Twee River fish. upstream to prevent re-invasion. This would establish a sanctuary for the undescribed Krom River taxa. Krom River, Joubertina (Eastern Cape) The Krom River originates in the Formosa State Forest Environmental impact assessment and drains eastwards along the Langkloof near the eastern The use of piscicides is not a ‘listed activity’ in terms of boundary of the Western Cape. Two undescribed taxa South Africa’s environmental legislation and there was, African Journal of Aquatic Science 2012, 37(2): 131–142 139

Location of farm dams — putative source of 24°00ƍ E Micropterus salmoides in the upper Krom River Working for Wetlands Redfin weirs waterfall

Galaxias?? Redfin range Largemouth bass* Micropterus salmoides Galaxias Galaxias range waterfall Krom Bass range

Area requiring Redfin population Road to further survey Kareedouw R62

Cape galaxias Road Galaxias spp. River Krom River redfin 33°55ƍ S 01.53 km Pseudobarbus sp. “after Krom” 24°00ƍ E

Figure 5: Map of the Krom River (Eastern Cape) showing landmarks, barriers and fish distributions (* denotes non-native species)

therefore, no legal requirement to conduct an environ- Systematic monitoring was recommended for each of the mental impact assessment (EIA) for this project. However, four rivers before, during and after the treatments. The legal under the National Environmental Management Act assessment component of the EIA concluded that Section (RSA 1998a), Section 28, a duty of care is required when 28 of NEMA placed an obligation on CapeNature to remedy engaging in ‘activities that may have a detrimental effect the degradation of environments harmed by non-native and/ on the environment’. In addition, the National Water Act or invasive species and that the proposed project was in (RSA 1998b) requires that the impact of the introduction of accordance with international best practices for managing a chemical into a water body be formally assessed. Further, invasive non-native species (Enviro-Fish Africa 2009). the sponsor of the project, the World Bank, insisted that an The EIA concluded that the justification for the project EIA be completed to determine whether the proposed use and the choice of rivers was sound. In most of the rivers of piscicides is viable and environmentally safe in the CFR. proposed for the project, the use of piscicides, specifi- In 2008, CapeNature appointed Enviro-Fish Africa (Pty) cally those containing rotenone, was recommended. The Ltd to carry out the EIA on its behalf. The four rivers identi- Rondegat River was recommended as the first river to be Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 fied for the pilot project were selected on the basis that treated. In the upper part of the Krom River (Cederberg), (a) indigenous taxa were critically endangered and non-native the trial of physical eradication methods was recommended, fish were a major threat to their survival, (b) the eradication to minimise impacts on indigenous Clanwilliam rock catfish of non-native fish was feasible, (c) the rivers would provide and macroinvertebrates. Should the physical methods prove healthy habitat for the indigenous fishes when the non-native ineffective, rotenone treatment should proceed, with rescue species had been removed, and (d) the rivers were not of populations of aquatic fauna being kept in holding facilities major importance to anglers. for the duration of the treatment. The EIA concluded that the treatment would have some negative initial impacts on the aquatic invertebrate fauna but Angler perceptions and public education that the majority of organisms could be expected to survive During the EIA it became apparent that sectors of the the treatments. In addition, rapid recovery of the stream fly angling community (notably trout enthusiasts) took faunas was predicted, following colonisation from reaches exception to a river containing rainbow trout (i.e. the Krom up- and down-stream of the treatment reaches. Furthermore, River, Cederberg) being chosen for the project. Concern the project was unlikely to have any significant impacts, was also expressed regarding the use of piscicides and either positive or negative, on the regional conservation their impacts on non-target fauna, especially aquatic status of non-fish vertebrate fauna, because mammals, invertebrates. The media publicised the project and were reptiles, amphibians and birds would not be affected by the initially very critical of it, with negative articles appearing in concentrations of rotenone required to kill fish. The project newspapers, e.g. Cape Argus (Bamford 2008), magazines, was endorsed as vital for the survival of endangered fishes. e.g. Flyfishing (Thorpe 2008), Farmer’s Weekly (Steyn 140 Marr, Impson and Tweddle

2010), and on several fishing websites. The EIA played such a plan, progress towards indigenous fish conservation an important role in changing public attitudes towards the can be measured. Capacity for freshwater fish conservation project, because (1) the project was thoroughly assessed needs strengthening within CapeNature (Impson et al. by experienced independent specialists, and (2) the EIA 2002b, Impson 2007) and the Eastern Cape Department of held two rounds of public participation meetings, allowing Environmental Affairs and Eastern Cape Parks Board. stakeholders the opportunity to express their concerns and The profile of indigenous fishes needs to be raised in the have these answered by the specialists conducting the EIA. eyes of the general public and riparian landowners, with Even though the EIA recommended that the project greater stakeholder engagement, as conservation actions proceeds, strong opposition to the project still remains, tend to be directed towards taxonomic groups such as birds especially among angling groups (Marr 2012). Such groups that are perceived to be more worthy of conservation by the could potentially sabotage restoration projects involving alien general public (Czech et al. 1998). fish eradication by re-introducing the eradicated species or introducing new species. Anglers have been identified Acknowledgements — The authors acknowledge the insightful as one of the high-risk groups potentially responsible for comments of the anonymous reviewers. The project was initially future introductions of freshwater fishes (Richardson et al. generously funded by the Global Environmental Facility of the World Bank and managed as a dedicated CAPE project. The 2004) and have a long history of illegally introducing angling authors acknowledge the financial support of CAPE for the species (Cambray 2003a, 2003b, Lintermans 2004, Impson fieldwork conducted for this project. The contributions of CAPE 2006, Johnson et al. 2009). Ensuring their support for the Alien Fish Project managers including Leigh Potter and Louise project is essential to the long-term success of the project. Stafford are acknowledged. SMM acknowledges the financial CapeNature, through its Freshwater Angling Forum, is support of the DST/NRF Centre of Excellence for Invasion Biology working closely with local angling groups to improve relation- and the David and Elaine Potter Foundation during his PhD studies ships and increase awareness of fish conservation and and expresses his appreciation for the supervision of professors management issues in the Western Cape. The local angling Jenny Day, Charles Griffiths and Paul Skelton during the comple- groups include the Western Cape Bass Angling Association, tion of his PhD thesis. NDI would like to thank CapeNature for Cape Piscatorial Society, Western Cape Yellowfish Working generously providing work time since 2004 to focus on this project. He gratefully acknowledges the CAPE programme for funding the Group, Western Province Artlure Society, Bank Angling initial planning phase of the CAPE Alien Fish Project, and latterly Federation and the Federation of South African Flyfishers. Working for Water for supporting the final planning and implementation stages of the Rondegat project. DT led the environmental Project status in January 2012 impact assessment for EnviroFish Africa (EFA) and thanks EFA, In 2012, CapeNature intends to treat the lower Rondegat CapeNature, the South African Institute for Aquatic Biodiversity River with an approved piscicide, CFT Legumine. Two (SAIAB), and the general public for their active participation in the treatments will be applied over consecutive years in late EIA process. SAIAB, a National Facility of the National Research summer, with the possibility of more treatments if some Foundation (NRF), is thanked for its full support and use of facilities smallmouth bass survive the initial treatments. The use during the EIA process and in the preparation of this paper. The of at least two treatments is standard practice in the USA images of fish used in Figures 2–5 are reproduced from Atlas of Southern African Freshwater Fishes by LEP Scott, PH Skelton, AJ (BJ Finlayson, California Department of Fish and Game, Booth, L Verheust, J Doonley and R Harris © South African Institute 2010, pers. comm.). CapeNature has prepared guidelines for Aquatic Biodiversity (SAIAB), with permission. for the implementation of the project, including public involvement, communication, fish rescue, monitoring, References and treatment plans, in accordance with those adopted by the American Fisheries Society (Finlayson et al. 2000). Bamford H. 2008. Experts slam CapeNature’s plan to poison alien Comprehensive independent pre-treatment monitoring is fish in our rivers. Cape Argus May 10: 3. Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 currently being undertaken with post-treatment monitoring Barnard KH. 1943. Revision of the indigenous freshwater fishes of planned following the completion of the treatment. the S.W. Cape region. Annals of the South African Museum 36: 101–262. Concluding remarks Bomford M, Tilzey R. 1997. Pest management principles for European carp. In: Roberts J, Tilzey R (eds), Controlling carp: exploring options for Australia. Proceedings of a workshop 22–24 CapeNature’s piscicide project has undergone a compre- October 1996, Albury. Griffith, New South Wales: CSIRO Land hensive EIA and is now in the implementation phase. and Water. pp 9–20. This project is an important step towards management of Britton JR, Copp G, Brazier M, Davies G. 2011b. A modular non-native fish in the CFR and South Africa. If, as expected, assessment tool for managing introduced fishes according to risks the project succeeds in removing non-native fish from the of species and their populations, and impacts of management treated reaches with limited impact on other taxonomic actions. Biological Invasions 13: 2847–2860. groups, it will guide future eradication of non-native fish Britton JR, Gozlan RE, Copp GH. 2011a. Managing non-native fish from critical biodiversity areas. Management of non-native in the environment. Fish and Fisheries 12: 256–274. fish populations, however, goes beyond possessing tools Cambray JA. 2003a. The global impact of alien trout species – a review; with reference to their impact in South Africa. African for removal of non-native fish and needs to be part of Journal of Aquatic Science 28: 61–67. comprehensive conservation management plans. A draft Cambray JA. 2003b. Impact on indigenous species biodiversity conservation plan was proposed for the Western Cape caused by the globalisation of alien recreational freshwater fishes. (Impson et al. 2002a) but needs to be refined to include Hydrobiologia 500: 217–230. defined goals, responsibilities and accountability. Within Cambray JA. 2003c. The need for research and monitoring of the African Journal of Aquatic Science 2012, 37(2): 131–142 141

impacts of translocated sharptooth catfish, Clarias gariepinus, in (eds), Invasive species and biodiversity management. Dordrecht: South Africa. African Journal of Aquatic Science 28: 191–195. Kluwer Academic Publishers. pp 363–375. Chadderton WL. 2003. Management of invasive freshwater fish: Impson ND. 2006. Do fishermen know better?: the collapse of striking the right balance. In: Department of Conservation (ed.), outstanding bass fisheries in Theewaterskloof and Voelvlei dams Managing invasive freshwater fish in New Zealand. Proceedings and in the Breede River. Piscator 138: 54–62. of a workshop hosted by Department of Conservation, 10–12 Impson ND. 2007. Freshwater fishes. In: Cape Nature (ed.), Western May, 2001, Hamilton, New Zealand. Wellington: Department of Cape Province State of Biodiversity 2007. Stellenbosch: Cape Conservation. pp 71–73. Nature Scientific Services. pp 18–36. Clarkson RW, Marsh PC, Stefferud SE, Stefferud JA. 2005. Conflicts Impson ND. 2010. Indigenous fish stocking in the Ratels catchment. between native fish and nonnative sport fish management in the Piscator 142: 12–17. southwestern United States. Fisheries 30: 20–27. Impson ND, Bills IR, Cambray JA. 2002a. A conservation plan for Coke M. 1988. Freshwater fish conservation in South Africa: a rising the unique and highly threatened freshwater fishes of the Cape tide. Journal of the Limnological Society of southern Africa 14: Floristic Kingdom. In: Collares-Pereira MJ, Cowx IG, Coelho MM 29–34. (eds), Conservation of freshwater fishes: options for the future. Collares-Pereira MJ, Cowx IG. 2004. The role of catchment scale Oxford: Fishing News Books, Blackwell Science. pp 432–440. environmental management in freshwater fish conservation. Impson ND, Bills IR, Cambray JA. 2002b. Freshwater fishes. In: Fisheries Management and Ecology 11: 303–312. Western Cape Nature Conservation Board (ed.), Western Cape Cowling RM, Pressey RL. 2003. Introduction to systematic State of Biodiversity 2000. Stellenbosch: Western Cape Nature conservation in the Cape Floristic Region. Biological Conservation Conservation Board. pp 1–20. 112: 1–13. Impson ND, Hamman KCD. 2000. Conservation status, threats Cucherousset J, Olden JD. 2011. Ecological impacts of non-native and future prospects for the survival of freshwater fishes of freshwater fishes. Fisheries 36: 215–230. the Western Cape Province, South Africa. In: Cowx IG (ed.), Czech B, Krausman PR, Borkhataria R. 1998. Social construction, Management and ecology of river fisheries. Oxford: Fishing News political power, and the allocation of benefits to endangered Books, Blackwell Science. pp 418–427. species. Conservation Biology 12: 1103–1112. Impson ND, Marriott MS, Bills IR, Skelton PH. 2007. Conservation Darwall WRT, Smith KG, Tweddle D, Skelton P (eds). 2009. The biology and management of a critically endangered cyprinid, the status and distribution of freshwater biodiversity in Southern Twee River redfin, Barbus erubescens (Teleostei: Cyprinidae), Africa. Gland: IUCN and Grahamstown: South African Institute of the Cape Floristic Region, South Africa. African Journal of for Aquatic Biodiversity. Aquatic Science 32: 27–33. de Moor IJ, Bruton MN. 1988. Atlas of alien and translocated Impson ND, Tharme R. 1998. Translocation of indigenous indigenous aquatic animals in Southern Africa. South African freshwater fishes for conservation and flyfishing: the Boontjies National Scientific Programmes Report No. 144. Pretoria: Council River, Western Cape as a case study. Piscator 130: 87–92. for Scientific and Industrial Research. IUCN. 2000. IUCN guidelines for the prevention of biodiversity loss Driver A, Nel JL, Snaddon K, Murray K, Roux DJ, Hill L, Swartz ER, caused by alien invasive species. Available at http://www.iucn. Manuel J, Funke N. 2011. Implementation manual for freshwater org/themes/ssc/publications/policy/invasivesEng.htm [accessed ecosystem priority areas. WRC Report No. 1801/1/11. Pretoria: 19 June 2006]. Water Research Commission. Johnson BM, Arlinghaus R, Martinez PJ. 2009. Are we doing all Enviro-Fish Africa. 2009. Environmental impact assessment of we can to stem the tide of illegal fish stocking? Fisheries 34: the proposed eradication of invasive alien fishes from four river 389–394. sections in the Cape Floristic Region. Grahamstown: Enviro-Fish Lintermans M. 2004. Human-assisted dispersal of alien freshwater Africa (Pty) Ltd. fish in Australia. New Zealand Journal of Marine and Freshwater Finlayson BJ, Schnick RA, Cailteux RL, DeMong L, Horton WD, Research 38: 481–501. McClay W, Tichacek GJ, Thompson C. 2000. Rotenone use in Lowe SR, Woodford DJ, Impson ND, Day JA. 2008. The impact fisheries management: administrative and technical guidelines of invasive fish and invasive riparian plants on the invertebrate manual. Bethesda, Maryland: American Fisheries Society. fauna of the Rondegat River, Cape Floristic Region, South Africa. Hamman KCD, Thorne SC, Skelton PH. 1984. The discovery of the African Journal of Aquatic Science 33: 51–62.

Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 Cape Kurper, Sandelia capensis (Cuvier in C. & V., 1831) in the Marr SM. 2012. Conservation of the native freshwater fishes of the Olifants river system (Western Cape Province). The Naturalist Cape Floristic Region (South Africa): management of non-native 28: 24–26. species. PhD thesis, University of Cape Town, South Africa. Harrison AC. 1950. Introduction of exotic fishes to the Cape Marr SM, Sutcliffe LME, Day JA, Griffiths CL, Skelton PH. 2009. Province: Section one – the prelude. Piscator 16: 125–128. Conserving the fishes of the Twee River, Cederberg: revisiting Harrison AC. 1951a. Introduction of exotic fishes to the Cape the issues. African Journal of Aquatic Science 34: 77–85. Province: Section one – the prelude (continued). Piscator 17: Marsh PC, Pacey CA. 2005. Immiscibility of native and nonnative 22–25. species. In: Brouder MJ, Springer CL, Leon SC (eds), Restoring Harrison AC. 1951b. Introduction of exotic fishes to the Cape native fish to the lower Colorado River: interactions of native Province: Section two – trout. Piscator 17: 25–33. and non-native fishes. Alburquerque, New Mexico: US Fish Harrison AC. 1952a. Introduction of exotic fishes to the Cape and Wildlife Service and Boulder City, Nevada: US Bureau of Province: Section three – black bass and bluegills: largemouth Reclamation. pp 59–63. bass. Piscator 22: 57–64. Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent Harrison AC. 1952b. Introduction of exotic fishes to the Cape J. 2000a. Biodiversity hotspots for conservation priorities. Nature Province: Section three – black bass and bluegills: bluegills. 403: 853–858. Piscator 23: 92–95. Myers JH, Simberloff D, Kuris AM, Carey JR. 2000b. Eradication Hey SA. 1926. The rapture of the river: the autobiography of a revisited: dealing with exotic species. Trends in Ecology and South African fisherman [Reprinted 2006]. Johannesburg: Evolution 15: 316–320. Platanna Press. Nel JL, Murray KM, Maherry AM, Petersen CP, Roux DJ, Driver Huntley BJ. 1999. South Africa’s experience regarding alien A, Hill L, van Deventer H, Funke N, Swartz ER, Smith-Adao species: impacts and control. In: Sandlund OT, Schei PJ, Viken A LB, Mbona N, Downsborough L, Nienaber S. 2011. Technical 142 Marr, Impson and Tweddle

report for the National Freshwater Ecosystem Priority Areas Africa. South African Journal of Zoology 30: 71–81. project. WRC Report No. 1801/2/11. Pretoria: Water Research Steyn AJ. 2010. Turning back the clock. Farmer’s Weekly 18 June: Commission. 80–82. Picker M, Griffiths C. 2011. Alien and invasive animals: a South Swartz E, Tweddle D. 2011. Site scale conservation. In: Darwall African perspective. Cape Town: Struik Nature. WRT, Smith KG, Allen DJ, Holland RA, Harrison IJ, Brooks Pípalová I. 2006. A review of grass carp use for aquatic weed EGE (eds), The diversity of life in African freshwaters: under control and its impact on water bodies. Journal of Aquatic Plant water, under threat. An analysis of the status and distribution of Management 44: 1–12. freshwater species throughout mainland Africa. Cambridge, UK Richardson DM, Cambray JA, Chapman RA, Dean WRJ, Griffiths and Gland, Switzerland: IUCN. pp 277–282. CL, Le Maitre DC, Newton DJ, Winstanley TJ. 2004. Vectors and Thorpe P. 2008. Story of an African river: at risk and betrayed. pathways of biological invasion in South Africa – past, present, Flyfishing Magazine April/May: 27–29. and future. In: Ruiz GM, Carlton JT (eds), Invasive species: Tweddle D, Bills R, Swartz E, Coetzer W, Da Costa L, Engelbrecht vectors and management strategies. New York: Island Press. pp J, Cambray J, Marshall B, Impson D, Skelton P, Darwall WRT, 292–349. Smith KG. 2009. The status and distribution of freshwater fishes. RSA (Republic of South Africa). 1998a. National Environmental In: Darwall WRT, Smith KG, Tweddle D, Skelton P (eds), The Management Act (Act No. 107 of 1998). Government Gazette, status and distribution of freshwater biodiversity in southern South Africa 401(19519). Africa. Gland: IUCN and Grahamstown: South African Institute RSA (Republic of South Africa). 1998b. National Water Act (Act No. for Aquatic Biodiversity. pp 21–37. 36 of 1998). Government Gazette, South Africa 398(19182). van Rensburg BJ, Weyl OLF, Davies SJ, van Wilgen LJ, Peacock RSA (Republic of South Africa). 2003. Protected Areas Act (Act No. DS, Spear D, Chimimba CT. 2011. Invasive vertebrates of South 57 of 2003). Government Gazette, South Africa 464(26025). Africa. In: Pimentel D (ed.), Biological invasions: economic and RSA (Republic of South Africa). 2004. National Environmental environmental costs of alien plant, animal, and microbe species Management: Biodiversity Act (Act No. 10 of 2004). Government (2nd edn). Boca Raton, Florida: CRC Press. pp 326–378. Gazette, South Africa 467(26436). van Rensburg KJ. 1966. Die vis van die Olifantsrivier (weskus) met Saunders DL, Meeuwig JJ, Vincent CJ. 2002. Freshwater protected spesiale verwysing na die geelvis (Barbus capensis) en saagvin areas: strategies for conservation. Conservation Biology 16: (Barbus serra). Cape Provincial Administration, Department of 30–41. Nature Conservation, Research Report 10. Shelton JM, Day JA, Griffiths CL. 2008. Influence of largemouth Vinson MR, Dinger EC, Vinson DK. 2010. Piscicides and inverte- bass, Micropterus salmoides, on abundance and habitat selection brates: after 70 years, does anyone really know? Fisheries 35: of Cape galaxias, Galaxias zebratus, in a mountain stream in the 61–71. Cape Floristic Region, South Africa. African Journal of Aquatic Weaver R. 2008. The historical Jonkershoek hatchery. Piscator Science 33: 201–210. 140: 24–29. Skelton PH. 1983. Perspectives on the conservation of threatened Wishart MJ, Day JA. 2002. Endemism in the freshwater fauna of fishes in Southern Africa. The Naturalist 27: 3–12. the south-western Cape, South Africa. Verhandlungen der Skelton PH. 1987. South African Red Data Book – fishes. South Internationale Vereinigung für Limnologie 28: 1–5. African National Scientific Programmes Report No. 137. Pretoria: Woodford DJ, Impson ND. 2004. A preliminary assessment of the Council for Scientific and Industrial Research. impact of alien rainbow trout (Oncorhynchus mykiss) on indigenous Skelton PH. 1990. The conservation and status of threatened fishes fishes of the upper Berg River, Western Cape Province, South in southern Africa. Journal of Fish Biology 37 (Suppl. A): 87–95. Africa. African Journal of Aquatic Science 29: 109–113. Skelton PH. 2000. Flagships and fragments – perspectives on the Woodford DJ, Impson ND, Day JA, Bills IR. 2005. The predatory conservation of freshwater fishes in southern Africa. African impact of invasive alien smallmouth bass, Micropterus dolomieu Journal of Aquatic Science 25: 37–42. (Teleostei: Centrarchidae), on indigenous fishes in a Cape Skelton PH. 2001. A complete guide to the freshwater fishes of Floristic Kingdom mountain stream. African Journal of Aquatic Southern Africa (2nd edn). Cape Town: Struik. Science 30: 167–173. Skelton PH. 2002. An overview of the challenges of conserving Younge A, Fowkes S. 2003. The Cape Action Plan for the freshwater fishes in South Africa. In: Collares-Pereira MJ, Cowx Environment: overview of an ecoregional planning process.

Downloaded by [University of Stellenbosch] at 06:30 25 January 2013 IG, Coelho MM (eds), Conservation of freshwater fishes: options Biological Conservation 112: 15–28. for the future. Oxford: Fishing News Books, Blackwell Science. Zambrano L, Martinez-Meyer E, Menezes N, Peterson AT. 2006. pp 221–236. Invasive potential of common carp (Cyprinus carpio) and Nile tilapia Skelton PH, Cambray JA, Lombard A, Benn GA. 1995. Patterns of (Oreochromis niloticus) in American freshwater systems. Canadian distribution and conservation status of freshwater fishes in South Journal of Fisheries and Aquatic Sciences 63: 1903–1910.

Received 11 March 2011, accepted 7 February 2012