Available online at www.sciencedirect.com

South African Journal of Botany 83 (2012) 36–43 www.elsevier.com/locate/sajb

Species limits in (Acacia) karroo (: Leguminoseae): Evidence from automated ISSR DNA “fingerprinting” ⁎ C.L. Taylor, N.P. Barker

Department of Botany, Rhodes University, P.O. Box 94, Grahamstown 6140, South Africa

Received 26 April 2012; received in revised form 13 July 2012; accepted 16 July 2012 Available online xxxx

Abstract

Vachellia karroo (Hayne) Banfi & Galasso, previously known as Acacia karroo Hayne, is a very common woody species in South Africa, and displays a large amount of morphological and architectural variation. Previous morphological studies have provided evidence to support the recognition of a number of segregate species, but this has not been supported by early protein electrophoresis studies. Genetic variability of the species throughout South Africa was examined by means of the Inter-Simple Sequence Repeat (ISSR) DNA “fingerprinting” to determine whether there is any genetic structure that correlates to the morphological diversity. Based upon 33 samples from across the South African distribution of this species, we found no evidence for any genetic structuring, suggesting the species is one panmictic entity. The existence of morphologically discrete varieties in light of this genetic finding is difficult to explain, and perhaps the species ought to be considered as an ochlospecies. A scenario whereby the various morphotypes evolved in refugia during previous climates followed by subsequent expansion and introgression during the current interglacial may explain this conundrum. Alternatively, the evolution of the morphotypes has been recent and rapid, and the genetic variation observed here represents the ancestral gene pool that has not yet undergone lineage sorting as a consequence of isolation. © 2012 SAAB. Published by Elsevier B.V. All rights reserved.

Keywords: Acacia; Automated ISSR; DNA fingerprinting; Genetic variation; Morphological variability; Ochlospecies; Vachellia karroo

1. Introduction bushveld, woodland, grassland, river banks and coastal dunes (Ross, 1971a,b, 1975; Verdoorn, 1954). However, it is absent Vachellia (Acacia) karroo (Hayne) Banfi & Galasso, from higher altitudes, and has been previously described as a commonly known as “sweet thorn” or “soetdoring”(Afrikaans), “ring species” as its distribution encircles the Great Escarpment and “umuNga” (Zulu Xhosa), “Mookana” (north Sotho) or “Mooka” particularly Drakensberg region (Brain, 1989; Ward, 2011). (Tswana) is a leguminous shrub common throughout southern Vachellia karroo shows a huge variety in terms of its growth Africa, ranging from the Southwestern Cape northwards into form, with from different areas in the species' geographical Namibia, Angola, Botswana, Zambia and Zimbabwe. It is a range often having a different appearance. In the latest formal pioneer species and has the ability to encroach rapidly into taxonomic revision, Ross (1971a) detailed the vast range in grassland grazing areas, and V. karroo is now considered the morphology in this species, and later (Ross, 1975) described most important woody invader of grasslands in South Africa seven different informal taxonomic entities of V. karroo which (O'Connor, 1995). It is an exceedingly variable species and the were generally correlated to distribution. He commented that the most widespread Acacia in southern Africa, occupying a diverse extremes of each variant are linked to the central V. karroo gene range of habitats including dry thornveld, river valley scrub, pool (in the Karoo and drier parts of the Cape Province) by numerous and varied intermediate stages and that these stages ⁎ Corresponding author. Tel.: +27 46 6037398; fax: +27 46 6037355. become progressively less and less distinct until it becomes E-mail address: [email protected] (N.P. Barker). difficult to assign a specimen to a particular entity with any real

0254-6299/$ -see front matter © 2012 SAAB. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.sajb.2012.07.014 C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43 37 certainty. Vachellia karroo is thus as an “inherently variable phenotypic plasticity, a change in the environment will enable an species” (Ross, 1975) in which no formal infraspecific adaptive phenotype to become further adaptive through natural categories could be recognized. Later, Swartz (1982) in an selection, which would then improve the performance of the unpublished thesis used 38 morphological characters to population, resulting in the assimilation of that trait in the new examine the variation of V. karroo. Different populations of environment (Pigliucci et al., 2006) This process has been termed V. karroo were analyzed by computerized numerical taxonom- genetic assimilation (Lande, 2009; Pigliucci et al., 2006; Price et ic methods and six infraspecific groups were defined. al., 2003). Seedlings from the six groups retained their distinct morphol- Assuming these forms are the product of ongoing evolu- ogy when grown in cultivation indicating that the infraspecific tionary selection, it should be possible to identify genotypes groups' characteristics are genetically fixed. This latter finding that correspond to regional differences in morphology; the hasalsobeenreportedbyMboumba and Ward (2008) and hypothesis that genetic diversity will reflect morphological Ward (2011). diversity (genotypes reflect morphotypes). Here we report on The various growth forms have thus been recognized as the use of Inter-Simple Sequence Repeat (ISSR) DNA varieties or ecotypes and even separate species (Coates Palgrave, fingerprinting to assess genetic diversity across the distribution 2002; Hurter and Van Wyk, 2004; Ross, 1971a, 1975; Smit, 1999, range and morphotypes in V. karroo. 2008; Swartz, 1982, 2003), and the various segregate species' ISSR is a PCR based technique using relatively short primers names are now increasingly being used in the popular (e.g. Coates that flank microsatellite regions. The primers are based on a Palgrave, 2002; Smit, 1999, 2008 who recognize taxa such as microsatellite repeated sequence and terminate in one or more A. robertsii, A. dyeri, A. kosiensis and A. theronii [=A. montana]) 3′-anchor nucleotides that differ from the repeat sequence. The and scientific literature (e.g. Mucina and Rutherford, 2006,who method thus requires no previous sequence knowledge (Godwin recognize A. natalitia as a discrete species, as recommended by et al., 1997). The method has been found to be useful for assessing Von Breitenbach (1989)). Table 1 summarizes the taxonomic genetic diversity, and the method has been shown to usually entities previously recognized by Ross (1975) and subsequent detect a higher level of polymorphism than restriction fragment authors. The International names Index (IPNI, accessed length polymorphism (RFLP) or random amplified polymorphic online at http://www.ipni.org/index.html) indicates Coates DNA (RAPD) (Godwin et al., 1997), and a much higher Palgrave (2002) as the original publication for many of these efficiency in identifying genetic individuals than allozymes (Li names (Table 1) but the nomenclatural validity of at least some and Ge, 2001). In addition, the ISSR technique has previously of these names is questionable, as we were unable to locate been found to be repeatable (Fang and Roose, 1997; Paterson et validly published formal Latin diagnoses and typifications for al., 2009; Taylor et al., 2011). The ISSR technique samples a wide these names. region of the genome and is well suited to population level genetic studies (Abbot, 2001; Paterson et al., 2009) and for investigating 1.1. Previous genetic diversity studies genetic variation within species (Ge et al., 2005; Zietkiewicz et al., 1994). The method has been successfully used in studies on Early work exploring the genetic diversity in this species by genetic diversity of rare plants (Ge et al., 2005; Xiao et al., 2004), Brain (1986) found that the expression of the different alleles of invasive weeds (Paterson et al., 2009) and a variety of crops the peroxidise enzyme was associated with geographic region. It (Amel et al., 2004; Ammiraju et al., 2001; Fang and Roose, 1997; was further demonstrated that there was an association of the Mondal, 2002). different alleles with different climatic variables (Brain, 1986), suggesting that populations from different regions have become 2. Methods locally adapted and may represent distinct evolutionary lineages. A subsequent peroxidise study based on over 3000 individuals 2.1. Testing the integrity of silica-dried leaf material showed that there were three genetic races: the “Western Karroo”, the “Eastern Cape” and the “Natal-Lowveld” races (Brain, 1989). As collecting fresh material and transporting it great distances However a later study which included a wider range of samples to the laboratory is not feasible, rapidly drying leaf material using did not support the earlier results (Brain et al., 1997), highlighting silica gel has been routinely used and is reported not to result in the need to sample the entire distribution range. much degradation of the DNA (Chase and Hills, 1991). However, Mboumba and Ward (2008) compared phenotypic plasticity in this is seldom tested, especially in the range of so-called “DNA two extreme populations of V. karroo growing 1300 km apart and fingerprinting” applications. However, if the treatment of silica investigated genetic variation among the populations by means of drying negatively affects the quality of the DNA, the ISSR allozyme markers. Their results suggested genetic differentiation in analysis will not give a complete or reliable banding pattern when plastic responses between populations, as the interactions between compared to DNA extracted from fresh material. Although dried the environment and some traits were not purely caused by the material has been used in ISSR analysis (e.g. Archibald et al., environment. Both Archibald and Bond (2003) and Mboumba and 2006; Bergh et al., 2007; Ge et al., 2005; Xia et al., 2007, 2006), Ward (2008) give evidence for phenotypic plasticity as a possible DNA extraction from old plant collections may produce a lower mechanism for changes in V. karroo. Ward (2011) proposes that yield with reduced quality (Hassel and Gunnarsson, 2003). Prior adaptive phenotype plasticity followed by genetic assimilation is to embarking on a large scale ISSR survey, we tested the effect the means by which differentiation occurs. If a plant shows that silica gel drying has on DNA extract and subsequent ISSR 38 C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43

Table 1 Summary of the various infraspecific forms of Vachellia (Acacia) karroo recognized by Ross (1975, 1979) and later workers. Ross (1975, 1979) Swartz (1982) Coates Palgrave (2002) Smit (2008) Acacia karroo (“typical” form) Rough-leaved variant Acacia karroo Acacia karroo (“growweblaarvariasie”) White-barked trees or shrubs with short spines, Fine-leaved variant Acacia natalitia E. Mey. Acacia natalitia E. Mey. 4—7 (13) pinnae with 12—18(27) pairs of leaflets (“fynblaarvariasie”) (A. natalitia); E. Cape, KZN, Swaziland, Mpumalanga Small, slender shrubs up to 1 m high; E. Cape Shrubby variant Acacia dyeri P.P. Swartz Acacia dyeri P.P. Swartz (Kei River Mouth) (“struikvormvariasie”) (noted under entry for Acacia karroo) “Fire resistant” shrubs from Nongoma district, KZN (A. inconflagrabilis) “Spindle A. karroo”; slender, sparingly Slender variant Acacia montana P.P. Swartz =Acacia theronii P.P. Swartz branched trees up to 6 m high; northern (“langgroivormvariasie”) (=A. montana P.P. Swartz; Swartz, 2003) KZN & Loskop dam region. Large trees with greyish bark and long spines (25 cm long) Coastal dune variant Acacia kosiensis P.P. Swartz Acacia kosiensis P.P. Swartz and long moniliform pods; KZN coast north of (“kusduinvariasie”) Tugela River into Mozambique. Form with sparse indumentum on young branchlets, leaves, Pubescent variant Acacia robbertsii P.P. Swartz Acacia robbertsii P.P. Swartz peduncles and pods (A. karroo var. transvaalensis); (“harige variasie”) Pretoria eastwards; similar to A. gerrardii. Small shrubby plants, similar to A. tenuispina, possibly hybrids with this species? Acacia sp. Acacia sekhukhuniensis P.J.H. Hurter (following Hurter and van Wyk, 2004) analysis using two sets of leaves collected from seven trees from a quantity of Taq DNA polymerase enzyme. The final ISSR PCR population near the Ecca Pass in the Eastern Cape (33°12′02″ S protocol used a concentration of 2.5 mM MgCl2, and the 26°37′06″E). DNA was extracted from one of the sets while the thermal cycling parameters comprised a 2 min denaturing step at leaves where still fresh and the other set was dried in silica gel for 94 °C, followed by 35 cycles of 94 °C for 30 s, 44 °C for 45 s and a week. DNA was then extracted from the dried leaf material. 72 °C for 90 s and a final extension of 20 min at 72 °C. The All DNA extractions for this experiment used the standard products of the PCR were electrophoresed on a 1% agarose gel and CTAB genomic DNA isolation technique (Doyle and Doyle, viewed using a UV transilluminator to determine the suitability of 1987) and not the commercially available kit used for the main the primers. Two primers were identified as being most suitable in ISSR survey. ISSR analyses were conducted on the DNA that they produced the most polymorphic bands (Table 3). extracted from both the dried and fresh material from the same These were Primer 864 (5′-ATGATGATGATGATGATG-3′) plant, and the fragment profiles compared. and Primer 812 (5′-GAGAGAGAGAGAGAGAA-3′). These primers were then manufactured (by Applied 2.2. Sampling and DNA extraction Biosystems) to include a fluorescent label which enabled the automatic detection of PCR products using a ABI 3130 genetic Thirty two samples of Vachellia karroo were collected analyzer with an automated detection system (Central Analytical from various sites throughout South Africa (Fig. 1A), covering Facility, Stellenbosch University). As noted by Archibald et al. the distributions ranges of at least two of the species segregates (2006), the increased sensitivity of the automated detection of (V. karroo s.s.andV. natalitia; Table 2; Fig. 1A). Leaf samples ISSR fragments results in much larger datasets than other scoring where dried in silica gel. In order to be consistent and to achieve methods. Primer 864 was labelled with NED and Primer 812 with good quality extracts, DNA was extracted using the Qiagen PET. Resultant electropherograms were analyzed using Peak DNeasy® Plant Tissue kit (Qiagen Inc., Valencia, CA) following Scanner Software version 1.0 (Applied Biosystems) with a the manufacturers protocol. medium level of peak smoothing. The dataset was exported for scoring in RawGeno ver. 2 (Arrigo et al., 2009), an automated 2.3. ISSR-PCR amplification and analysis DNA fragment scoring application using the R ver. 2.11.1 platform (The R Foundation for Statistical Computing). PCR Thirteen universal ISSR primers (designed by Wolfe and reactions which were deemed to be of poor quality (compara- Ballard; Wolfe et al., 1998; McCauley and Ballard, 2002) were tively few bands detected) were filtered by removing individuals tested to determine which were suitable for further use in this with low band replication; the 5% of individuals with the lowest study. Suitable primers were those that amplified high numbers and highest number of bands were removed before the analysis. of products easily and which showed consistency of amplifi- Additional filtering parameters for scoring by RawGeno were bin cation in repeated PCR reactions. A series of PCR reactions width, which was left as the default (minimum: 1; maximum: 1.5) were also undertaken to optimise the ISSR PCR protocol, and the scoring range, which was from 100 base pairs to 650 base especially in terms of magnesium chloride concentration and pairs. C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43 39

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 11 25 -17 -17 15 -18 -18 29 13 -19 A -19 B 23 21 -20 -20 38 -21 -21 16 864 22 -22 -22 24 26 -23 -23 12 -24 -24 17 28 -25 -25 33 40 -26 -26 35 -27 -27 19 27 -28 -28 31 -29 -29 44 45 -30 -30 32 30 -31 -31 34 -32 -32 42 39 -33 -33 36 41 -34 -34 37

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 0.68 0.74 0.80 0.86 0.92 Coefficient 11 12 11 C 14 D 12 39 28 15 15 18 864 + 812 26 812 42 29 32 22 34 23 26 16 30 24 45 30 35 45 23 22 32 29 42 24 34 27 35 28 19 16 39 19 27 37 37 0.66 0.72 0.78 0.84 0.90 0.66 0.70 0.75 0.79 0.83 Coefficient Coefficient E F 39 0.1 30 45 0.27 28 19 32 42 12 34 24 27 0.14 16 29 22 35 11 37 23 45 26 11 37 34 30 15 32 0.01 42 35 12 16 28

26 Axis 2 8.16% 15 25 -0.12 29 27 22 23 24

39 -0.25 -0.50 -0.35 -0.20 -0.05 0.10 Axis 1 20.43%

Fig. 1. (A) Map of sampling localities. Dots represent samples amplified using ISSR primer 864 only, triangles represent samples amplified using both primers (864 and 812), and sample numbers correspond to numbers appearing in Table 2 and B–F; (B) UPGMA phenogram obtained from samples amplified with primer 864 and a fluorescence cut-off value of 100; (C) UPGMA phenogram obtained from samples amplified with primer 812 only and a fluorescence cut-off value of 100; (D) UPGMA phenogram obtained from subset of samples amplified with both primers and a fluorescence cut-off value of 100; (E) Neighbor-net analyses of combined data set (primers 812 and 864); (F) Principal Components Analysis (PCA) of combined data set (primers 812 and 864). 40 C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43

Table 2 Localities of Vachellia karroo samples. Sample number refers to numbers shown in Fig. 1. Key to collectors: IM=I. Mackellar, LL=Linda Loffler, MV=Martin Villet, NPB=Nigel Barker, RD=Ryan Daniels., RMK=Robert McKenzie, SW=S. Walton, VDM=V. van der Merwe. ISSR primer Sample No. Longitude (S) Latitude (E) Province Collector Identity 812 864 11 −24.1482 27.77391 Limpopo VDM 123 Vachellia karroo XX 12 −23.2936 27.76472 Limpopo VDM 131 Vachellia karroo XX 13 −29.12527 19.40498 Northern Cape VDM 180 131 Vachellia karroo X 14 −22.0027 27.81515 Botswana VDM 4 Vachellia karroo X 15 −23.0565 29.49432 Limpopo VDM 15 Vachellia karroo XX 16 −24.1496 30.21423 Limpopo VDM 22 Vachellia karroo XX 17 −24.5055 30.84482 Limpopo VDM 29 Vachellia karroo X 18 −27.8236 31.42030 Kwa-Zulu Natal VDM 34 Vachellia karroo X 19 −32.2600 28.77250 Eastern Cape NPB 2061 Vachellia karroo XX 21 −26.0747 31.18472 Swaziland LL s.n. Vachellia natalitia X 22 −34.0714 22.1608 Western Cape RMK 2589 Vachellia karroo XX 23 −32.4612 24.08694 Eastern Cape RMK 2615 Vachellia karroo XX 24 −32.2814 24.55514 Eastern Cape RMK 2616 Vachellia karroo XX 25 −33.0871 21.57864 Western Cape RMK 2596 Vachellia karroo X 26 −34.1524 21.30067 Western Cape RMK 2590 Vachellia karroo XX 27 −23.7165 27.74229 Limpopo VDM 129 Vachellia karroo XX 28 −22.622 28.90578 Limpopo VDM 145 Vachellia karroo XX 29 −27.9072 25.2697 North West VDM 164 Vachellia karroo XX 30 −30.3713 17.49904 Northern Cape VDM 181 Vachellia karroo XX 31 −23.7126 29.99521 Limpopo VDM 17 Vachellia karroo X 32 −27.7510 31.16234 Kwa-Zulu Natal VDM 38 Vachellia natalitia XX 33 −26.4461 31.18694 Swaziland LL s.n. Vachellia natalitia X 34 −25.7558 28.2180 Pretoria IM s.n. Vachellia karroo XX 35 −33.0552 26.65389 Eastern Cape NPB 2011 Vachellia karroo XX 36 −33.1908 26.61972 Eastern Cape NPB 2008 Vachellia karroo X 37 −33.1269 26.61444 Eastern Cape NPB 2009 Vachellia karroo XX 38 −32.7122 27.18417 Eastern Cape NPB 2023 Vachellia karroo X 39 −30.8189 30.39250 Kwa-Zulu Natal SW 30 Vachellia natalitia XX 40 −30.6031 30.54694 Kwa-Zulu Natal SW 32 Vachellia natalitia X 41 −32.9795 23.71089 Eastern Cape MV s.n. Vachellia karroo X 42 −33.8947 25.14944 Eastern Cape NPB 2024 Vachellia karroo XX 44 −33.5363 21.69139 Western Cape RD s.n. Vachellia karroo X 45 −33.7791 20.93261 Western Cape RD s.n. Vachellia karroo XX

Two data sets were created, one where the low florescence coefficient were generated in NTSYS ver. 1.40 (Rohlf, 2002). threshold was set at 100 florescence units and another where In addition, because infraspecific analyses may not reflect a the threshold was set to a less sensitive 200 florescence units. hierarchical structure (Huson and Bryant, 2006; Vriesendorp This was done to test whether samples that lacked or had only and Bakker, 2005), neighbor-net analyses were undertaken low intensity bands affected the result. The resulting binary using SplitsTree ver. 4.10. (Huson, 1998). matrix for each primer and each florescence parameter was exported as a tab-delimited text file and matrices were consolidated 3. Results in Microsoft® Excel. Results from each primer were analyzed separately as well as in a combined dataset. A Principle 3.1. Comparison of DNA extractions from fresh and dried Component Analysis (PCA) and a UPGMA phenogram based leaves on a matrix of similarities created using the Simple Matching The results from this preliminary test indicated that there were some differences between amplifications of DNA extracted from Table 3 Summary of bands generated from ISSR primers 812 and 864. fresh leaves and silica dried leaves (i.e. fragment profiles were not identical) However, phenetic analyses resolved each sample pair Primer Fluorescence Total no. Average no. bands No. private cut-off bands per sample bands (1 fresh and 1 dried) as most similar to each other (results not shown). It thus appears that drying the leaves using silica gel may 812 100 383 82 59 812 200 148 38 27 not result in ISSR profiles that completely match those obtained 864 100 294 54 50 using fresh material, but does produce sufficiently similar results 864 200 201 44 26 to enable the pairing of fresh and dried samples from the same Combined 100 655 141 120 population; samples which are presumed to be genetically Combined 200 338 86 53 similar. As we could not obtain fresh material from all the C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43 41 field-collected samples, for the sake of consistency we opted (Kuechly et al., 2010), and it is thus possible for these animals to to use only dried material in our study. also act as dispersal agents for V. karroo. Woody species tend to maintain more variation within species 3.2. Genetic variability within Vachellia karroo and within populations than other species but have less variation among populations (Hamrick et al., 1992). In addition to this, Analyses were conducted on both primers at a cut off of 100 woody species with large geographic ranges have more genetic and 200 inflorescent units. Primer 812 gave 383 bands and 148 diversity within species and populations but less variation among bands at 100 and 200 florescent units respectively, while primer populations compared to other more isolated woody species 864 gave 294 and 201 bands at 100 and 200 florescent units (Hamrick et al., 1992). Vachellia karroo is a widespread woody respectively (Table 3). The average number of bands and the species and therefore this might explain the lack of variation number of bands unique to a particular individual are also among populations. In addition to life-history and ecological summarized in Table 3. characteristics of woody species, another factor that plays an Although some differences were found between results from important role in determining the level and distribution of genetic the 100 florescent unit data set and the 200 florescent unit data diversity is the evolutionary history of the species (Hamrick et al., set for each individual primer, both data sets show limited 1992). A species that historically consisted of large continuous resolution in terms of groupings of samples, irrespective of populations would be less susceptible to the random loss of analytical method used (results not shown). The phenograms genetic variation. Therefore many different factors play signifi- derived from the individual data sets using the 100 fluorescent cant roles in shaping the present-day genetic structure of a species unit cut-off show no patterns or clusters (Fig. 1B, C). Where and its populations (Hamrick et al., 1992). samples were amplified for both primers, the data was merged into a combined data set comprising 20 samples. Phenetic 4.1. Taxonomic considerations analyses of this data set (which had 748 characters) also showed no clear resolution of any groupings or clusters irrespective of The different morphological forms of V. karroo are the analytical method. (Fig. 1D). Similarly, Neighbor Net and PCA product of ongoing evolution and natural selection. There is fail to resolve any discrete clusters of samples (Fig. 1EandF evidence that phenotypic plasticity is the mechanism for changes respectively). Eigen values of the PCA are low, the first two axes in V. karroo to environmental conditions (Archibald and Bond, accounting for only 20.43% and 8.16% of the variation, 2003; Mboumba and Ward, 2008). Ward (2011) proposed that suggesting that there is little signal in the data. Analyses of the adaptive phenotypic plasticity followed by genetic assimilation is 200 florescent unit cut-off dataset (which was reduced to 370 the means by which differentiation in morphology occurs. characters) produced very similar results, and hence these are not However, at the genetic level, genotypes that correspond to presented. regional differences in morphology and physiology could not be identified and it can be concluded that genetic differentiation does 4. Discussion not reflect the morphological differentiation observed. Vachellia karroo hasbeenconsideredasa“ring species” by The results from the ISSR analysis indicate that there is no some (Ward, 2011 and refs therein), having a circular distribution genetic clustering of samples within the V. karroo species that around the Drakensberg and allied mountains of the Great could be interpreted as supporting the existence of morphological Escarpment. However, the species complex could equally be forms or segregate species. This finding is thus in agreement with viewed as an ochlospecies. This species concept, first used by that of Brain et al. (1997), but in contrast to the distinct groups White (1998), describes a highly variable or polymorphic species found by Mboumba and Ward (2008), where the PCA of allozyme with chaotic infraspecific variation which is only partly correlated data revealed little genetic overlap. However, this latter study only with ecology and geography and which is of such a complex compared two populations of V. karroo and these populations were pattern as to be intractable to formal taxonomic treatment (Cronk, from two different extremes. This study on the other hand looked at 1998). Cronk (1998) lists 10 traits of an ochlospecies, and samples from different populations throughout South Africa and considers that if a species meet six or more of these then it may be therefore would have sampled varieties that may fall between two considered an ochlospecies. These traits are: extreme forms. The limited genetic differentiation could indicate that the 1. Non-hierarchical polymorphic variation; species dispersed relatively recently and thus no genetic differen- 2. Character-state distribution that is only partially correlat- tiation as such has occurred. Alternatively there may be continuous ed with geography and ecology; gene flow throughout the distribution of A. karroo thus accounting 3. Characters that vary independently and not in a correlated for the lack of differentiation and close relatedness between fashion; samples. Extensive gene flow could be both pollen and seed 4. Complexity of variation is not due to hybridization between mediated. Documented dispersal agents of seeds of this species currently recognisable species; include birds which use the plant material in their nests (Dean et al., 5. Geographically and ecologically widespread, occurring 1990) and ungulates such as domestic cattle and especially kudu in several climatic zones; (Miller, 1996). In addition, rodents have been shown to be effective 6. At a particular locality two distinct and non-intergrading seed dispersal agents in other southern African Vachellia species forms may be found, and other forms may be found at 42 C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43

other localities, but taken together all the forms intergrade References and the classification breaks down; 7. Have closely related, but morphologically distinct and Abbot, P., 2001. Individual and population variation in invertebrates revealed monotypic satellite species; by Inter-simple Sequence Repeats (ISSRs). Journal of Insect Science 1 (8). Amel, S.-H., Mokhtar, T., Salwa, Z., Jihe`ne, H., Messaoud, M., Abdelmajid, 8. Similar variants may occur in widely separate localities R., Mohamed, M., 2004. Inter-Simple Sequence Repeat fingerprints to and appear to be polytypic in origin; assess genetic diversity in Tunisian fig (Ficus carica L.) germplasm. 9. There is often a proliferation of synonyms; and Genetic Resources and Crop Evolution 51, 269–275. 10. Tendency for ochlospecies to occur in medium to large Ammiraju, J.S.S., Dholakia, B.B., Santra, D.K., Singh, H., Lagu, M.D., genera usually with more than 50 species. Tamhankar, S.A., Dhaliwal, H.S., Rao, V.S., Gupta, V.S., Ranjekar, P.K., 2001. Identification of inter simple sequence repeat (ISSR) markers associated with seed size in wheat. Theoretical and Applied Genetics 102, 726–732. Vachellia karroo can certainly meet traits 1, 5, 6 (Ward (2011) Archak, S., Gaikwad, A.B., Gautam, D., Rao, E.V.V.B., Swamy, K.R.M., reports the co-existence of two forms), 7, 8 and 10. With further Karihaloo, J.L., 2003. Comparative assessment of DNA fingerprinting study it is almost certain that this species can be found to meet techniques (RAPD, ISSR and AFLP) for genetic analysis of cashew (Anacardium occidentale L.) accessions of . Genome 46, 362–369. additional requirements for consideration as an ochlospecies. Archibald, S., Bond, W.J., 2003. Growing tall vs growing wide: tree architecture In summary, while discrete morphological entities can be and allometry of Acacia karroo in forest, savannah, and arid environments. identified within a broad Vachellia karroo concept (some of Oikos 102, 3–14. which have been elevated to species status) we feel this Archibald, J.K., Crawford, D.J., Santos-Guerra, A., Mort, M.E., 2006. The recognition is premature, and represents human efforts to utility of automated analysis of inter-simple sequence repeat (ISSR) loci for resolving relationships in the canary Island species of Tolpis (Asteraceae). place artificial hierarchical structure on an entity that has no American Journal of Botany 93, 1154–1162. such structure. It is worth emphasizing Ross's (1975, p. 94) Arrigo, N., Tuszynski, J.W., Ehrich, D., Gerdes, T., Alvarez, N., 2009. insightful comment that “… extremes of these variants are Evaluating the impact of scoring parameters on the structure of intra- usually distinctive and attract immediate attention. However, specific genetic variation using RawGeno, an R package for automating the extremes are linked to the ‘central’ A. karroo genepool by AFLP scoring. BMC Bioinformatics 10 (33), http://dx.doi.org/10.1186/ …” 1471-2105-10-33. numerous and varied intermediate stages The recognition Bergh, N.G., Hedderson, T.A., Linder, H.P., Bond, W.J., 2007. Palaeoclimate- of satellite species is thus only a consequence of recognising induced range shifts may explain current patterns of spatial genetic variation in the forms at the extremes of the ranges of morphological renosterbos (Elytropappus rhinocerotis, Asteraceae). Taxon 56, 393–408. variation, and ignoring the intermediates. While we have not Brain, P., 1986. Leaf peroxidase types in Acacia karroo. Geographical sampled the full range of the morphological variation in the distribution and influence of the environment. South African Journal of Botany 52, 47–52. species (including many of the species segregates recognized Brain, P., 1989. Genetic races in a ring species, Acacia karroo. South African by some taxonomists), our genetic data certainly corroborates Journal of Science 85, 181–185. Ross's interpretation (and Brain et al.'s, 1997 findings), as Brain, P., Harris, S.A., Barnes, R.D., 1997. Leaf peroxidase types in Acacia karroo there is no evidence for any segregation of samples based on Hayne (Acacieae, Leguminosae): a range-wide study. Silvae Genetica 46, 88–94. geographic distribution or the morphological forms included Chase, M.W., Hills, H.H., 1991. Silica gel: an ideal material for field preservation of leaf samples for DNA. Taxon 40, 215–220. here. Coates Palgrave, K., 2002. Trees of Southern Africa. Struik Publishers, Cape While samples for this study have been collected from across Town. South Africa, samples from both further afield and the complete Cronk, Q.C.B., 1998. The ochlospecies concept. In: Huxley, C.R., Lock, J.M., range of species segregates are required to further test our Cutler, D.F. (Eds.), Chorology, and ecology of the floras of findings. In addition, the use of other fingerprinting methods such Africa and Madagascar. Royal Botanic Gardens, Kew. Dean, W.R.J., Milton, S.J., Siegfried, W.R., 1990. Dispersal of seeds as nest as amplified fragment length polymorphisms (AFLP) markers material by birds in semi-arid Karoo shrubland. Ecology 71, 1299–1306. (which have a superior utility and a higher reliability and Doyle, J.J., Doyle, J.L., 1987. A rapid DNA isolation procedure for small repeatability — and considerably greater cost) compared to quantities of fresh leaf tissue. Phytochemistry Bulletin 19, 11–15. ISSR analyses (Archak et al., 2003) could also be used. Fang, D.Q., Roose, M.L., 1997. Identification of closely related citrus cultivars with inter-simple sequence repeat markers. Theoretical and Applied Genetics 95, 408–417. Ge, X.-J., Zhang, L.-B., Yuan, Y.-M., Hao, G., Chiang, T.-Y., 2005. Strong Acknowledgments genetic differentiation of the East-Himalayan Megacodon stylophorus (Gentianaceae) detected by Inter-Simple Sequence Repeats (ISSR). The DSF-NRF Centre for Excellence in Tree Heath Biodiversity and Conservation 14, 849–861. Biotechnology is gratefully acknowledged for financial support Godwin, I.D., Aitken, E.A.B., Smith, L.W., 1997. Application of inter simple sequence repeat (ISSR) markers to plant genetics. Electrophoresis 18, in the form of an Honours Bursary to CLT and running grant to 1524–1528. NPB. The Central Analytical Facility, Department of Genetics, Hamrick, J.L., Godt, M.J.W., Sherman-Broyles, S.L., 1992. Factors influencing Stellenbosch University is acknowledged for their assistance with levels of genetic diversity in woody plant species. New Forests 6, 95–124. the automated detection of fluorescently labelled PCR products. Hassel, K., Gunnarsson, U., 2003. The use of inter simple sequence repeats – We thank Dr. Iain Paterson and Ms. Sarah Taylor for their advice (ISSR) in bryophyte population studies. Lindbergia 28, 152 157. Hurter, P.J.H., Van Wyk, A.E., 2004. A new species of Acacia (Mimosoideae) with extractions and analysis, Ms. Bianca Wernecke for help with from the province of Limpopo, South Africa. Bothalia 34, 109–112. GIS, Dr. Sue Milton-Dean for her extensive literature base on Huson, D.H., 1998. SplitsTree: a program for analyzing and visualizing Vachellia (Acacia) karroo, and all the collectors listed in Table 2. evolutionary data. Bioinformatics 14, 68–73. C.L. Taylor, N.P. Barker / South African Journal of Botany 83 (2012) 36–43 43

Huson, D.H., Bryant, D., 2006. Application of phylogenetic networks in Smit, N., 1999. Guide to the Acacias of South Africa. Briza Publications, evolutionary studies. Molecular Biology and Evolution 23, 254–267. Pretoria, South Africa. Kuechly, H.U., Mueller, J.S., Reinfelder, V.L., Wiedemann, S., Blaum, N., Smit, N., 2008. Field Guide to the Acacias of South Africa. Briza Publications, 2010. Rodent-mediated dispersal of Acacia seeds in Kalahari savannah Pretoria, South Africa. rangelands — implications for bush encroachment. African Journal of Swartz, P.P., 1982. 'n Numeries-taksonomiese evaluering van die takson Acacia Ecology 49, 119–121. karroo Hayne in suidelike Afrika. (A numerical taxonomic evaluation of the Lande, R., 2009. Adaptation to an extraordinary environment by evolution of taxon Acacia karroo Hayne in southern Africa). MSc Thesis, University of phenotypic plasticity and genetic assimilation. Journal of Evolutionary Pretoria, Pretoria, South Africa. Biology 22, 1435–1446. Swartz, P.P., 2003. The correct name for Acacia montana. Bothalia 33, 164–165. Li, A., Ge, S., 2001. Genetic variation and clonal diversity of Psammochloa villosa Taylor, S.J., Paterson, I., Downie, D., 2011. Genetic diversity of introduced (Poaceae) detected by ISSR markers. Annals of Botany 87, 585–590. populations of the water hyacinth biological control agent Eccritotarsus Mboumba, G., Ward, D., 2008. Phenotypic plasticity and local adaptation in catarinensis (Hemiptera: Miridae). Biological Control 58, 330–336. two extreme populations of Acacia karroo. African Journal of Range and Verdoorn, I.C., 1954. The nomenclature of the Cape Acacia. Bothalia 6, Forage Science 25, 121–130. 409–413. McCauley, R.A., Ballard Jr., H.E., 2002. Inferring nativity and biogeographic Von Breitenbach, F., 1989. The lost Natal Thorn: For a reinstatement of Acacia affinities of central and marginal populations of Froelichia floridana natalitia E. Mey. Journal of Dendrology 12, 1–10. (Amaranthaceae) from Inter-Simple Sequence Repeat (ISSR) markers. Vriesendorp, B., Bakker, F.T., 2005. Reconstructing patterns of reticulate Journal of the Torrey Botanical Society 129, 311–325. evolution in angiosperms: what can we do? Taxon 54, 593–604. Miller, M.F., 1996. Dispersal of Acacia seeds by ungulates and ostriches in an Ward, D., 2011. Population differentiation in a purported ring species, Acacia African Savanna. Journal of Tropical Ecology 12, 345–356. karroo (Mimosoideae). Biological Journal of the Linnean Society 104, Mondal, T.K., 2002. Assessment of genetic diversity of tea (Camellia sinensis 748–755. (L.) O. Kuntze) by inter-simple sequence repeat polymerase chain reaction. White, F., 1998. The vegetative structure of African Ebenaceae and the Euphytica 128, 307–315. evolution of rheophytes and ring species. In: Fortune-Hopkins, H.C., Mucina, L., Rutherford, M.C. (Eds.), 2006. The vegetation of South Africa, Huxley, C.R., Pannell, C.M., Prance, G.T., White, F. (Eds.), The biological Lesotho and Swaziland. Strelitzia, 19. South African National Biodiversity monograph: the importance of field studies and functional syndromes for Institute, Pretoria. taxonomy and evolution of tropical plants. Royal Botanical Gardens, Kew. O'Connor, T.G., 1995. Acacia karroo invasion of grassland: environmental and Wolfe, A.D., Xiang, Q.Y., Kephart, S.R., 1998. Assessing hybridization in natural biotic effects influencing seedling emergence and establishment. Oecologia populations of Penstemon (Scrophulariaceae) using hypervariable intersimple 214–223. sequence repeat (ISSR) bands. Molecular Ecology 7, 1107–1125. Paterson, I.D., Downie, D.A., Hill, M.P., 2009. Using molecular methods to Xia, T., Chen, S., Chen, S., Zhang, D., Zhang, D., Gao, Q., Ge, Z., 2007. ISSR determine the origin of weed populations of Pereskia aculeata in South analysis of genetic diversity of the Qinghai-Tibet Plateau endemic Rhodiola Africa. Biological Control 48, 84–91. chrysanthemifolia (Crassulaceae). Biochemical Systematics and Ecology Pigliucci, M., Murren, C.J., Schlichting, C.D., 2006. Phenotypic plasticity and 35, 209–214. evolution by genetic assimilation. Journal of Experimental Biology 209, Xiao, L.-Q., Ge, X.-J., Gong, X., Hao, G., Zheng, S.-I., 2004. ISSR Variation in 2362–2367. the Endemic and Endangered Plant Cycas guizhouensis (Cycadaceae). Price, T.D., Qvarnstrom, A., Irwin, D.E., 2003. The role of phenotypic Annals of Botany 94, 133–138. plasticity in driving genetic evolution. Proceedings of the Royal Society of Xiao, M., Li, Q., Wang, L., Guo, L., Li, J., Tang, L., Chen, F., 2006. ISSR London B, Biological Sciences 270, 1433–1440. analysis of the genetic diversity of the endangered species Sinopodophyllum Rohlf, F.J., 2002. NTSYSpc: Numerical Taxonomy System, Version 2.0. hexandrum (Royle) Ying from western Sichuan Province, China. Journal of Exeter Publishing, Ltd., Setauket, NY. Integrative Plant Biology 48, 1140–1146. Ross, J.H., 1971a. Acacia karroo in Southern Africa. Bothalia 10, 385–401. Zietkiewicz, E., Rafalski, A., Labuda, D., 1994. Genome fingerprinting by Ross, J.H., 1971b. A Variant of Acacia karroo from Sekukuniland (Eastern simple sequence repeat (SSR)-anchored polymerase chain reaction ampli- Transvaal). Bothalia 10, 427–430. fication. Genomics 20, 176–183. Ross, J.H. (Ed.), 1975. Flora of Southern Africa, vol. 16. Government Printer, Pretoria, South Africa (part 1).

Edited by JC Manning