Printer Friendly

African cycad ecology, ethnobotany and conservation: a synthesis.

Introduction

Globally, the cycads are a group of gymnosperms of ancient origin, comprising 344 species in three families and 10 genera (Calonje et al., 2016). Cycads are long-lived, slow-growing, perennial, dioecious plants, and occur in most of the world's tropical and subtropical regions (Jones, 1993; Osborne, 1995; Whitelock, 2002; Donaldson, 2003). Since their discovery by European naturalists and botanists in the 1700-1800s, cycads have attracted significant botanical and horticultural interest worldwide, and as a result, thousands of plants have been collected from the wild for private collections and botanical gardens (Pearson, 1905; Giddy, 1984; Osborne, 1995; Donaldson, 2003). Cycads are currently listed as the world's most threatened group of organisms, with 62% of the known species threatened with extinction (Hoffmann et al., 2010). Hence, despite their representation of only a small fraction of the world's plant diversity, they are a group of global conservation significance (Donaldson, 2003).

Africa hosts all three cycad families: the Cycadaceae, Stangeriaceae and Zamiaceae, each of which is represented on the continent by a single genus (Figs. 1 & 2) (Goode, 2001; Donaldson, 2003). Cycas thouarsii is the only African representative of Cycas, the sole genus in the Cycadaceae, comprising 113 species, all of which occur in Australasia except for C. thouarsii (Donaldson, 2003; Calonje et al., 2016). The Stangeriaceae consists of two genera: Bowenia (which comprises two species, both endemic to Australia) and Stangeria, which contains the single species S. eriopus and is endemic to South Africa and southern Mozambique (Jones, 1993; Whitelock, 2002; Donaldson, 2003). The Zamiaceae is the largest cycad family, consisting of seven genera: four occur in the Americas (Ceratozamia, Dioon, Microcycas and Zamia), two in Australia (Lepidozamia and Macrozamia), and Encephalartos is endemic to Africa (Jones, 1993; Whitelock, 2002; Donaldson, 2003). The African continent also harbours a global cycad diversity hotspot in South Africa, where 38 taxa occur (Donaldson, 2003, 2008).

African cycads have been the subject of several scientific studies, and, owing to their conservation significance, have also been the focus of numerous conservation action plans and monitoring programmes. Several books have been published on cycads, some of which have dealt with them as a group of plants in general, and included descriptions of African cycads (e.g., Chamberlain, 1919; Jones, 1993; Norstog & Nicholls, 1997; Whitelock. 2002), while others have focused exclusively on Africa's cycads (Goode, 1989, 2001), or on taxa from a particular region or country on the continent (e.g., Dyer, 1965; Heibloem, 1999; Giddy, 1984; Grobbelaar, 2004). While much of the key ecological information on the continent's cycads, including their fire survival, role in the ecosystem, resprouting ability, population dynamics and reproductive ecology, is found in the peer-reviewed literature, the remainder is scattered across popular journals, scientific reports, dissertations, theses and anecdotes. Currently there is no published synthesis of this information, making it difficult to access by researchers and conservation practitioners working on cycads. Hence, the aim of this paper is to review the literature on African cycad ecology with additional foci on the ethnobotany and conservation of the continent's cycads. Furthermore, the paper seeks to elucidate general trends and highlight opportunities for further study, thereby providing a resource to guide future research and conservation-related work on African cycads.

Geographical Distribution

Africa and the adjacent Indian Ocean islands harbour a total of 67 cycad species comprising 65 Encephalartos species, Stangeria eriopus and Cycas thouarsii (Fig. 1) (Goode, 2001; Donaldson, 2003; Calonje et al., 2016). Only three Encephalartos species--E. barteri, E. ferox and E. tegulaneus--are recognized as having subspecies (two each), bringing the total number of Encephalartos species and subspecies to 68 (Calonje et al., 2016). Africa's cycads occur mainly along the eastern side of the continent, from South Africa to Kenya, extending across central Africa (Democratic Republic of the Congo and Uganda) to Angola, Nigeria, Benin and Ghana (Goode, 1989, 2001; Jones, 1993; Donaldson, 2003) (Fig. 1). Cycads occur in 16 mainland African countries and the islands of Madagascar, Comoros, Seychelles and Zanzibar along the continent's east coast (Goode, 2001). Stangeria eriopus occurs only in South Africa (Eastern Cape and KwaZulu-Natal provinces) and southern Mozambique. It may be argued that Cycas thouarsii once occurred solely on Madagascar, being introduced to the African continent by humans during the heyday of mercantile activity in the western Indian Ocean (Goode, 1989). This species does, however, produce seeds that float in water, and could conceivably have been carried to the African coast by ocean currents (Goode, 1989).

Despite their broad geographic range throughout sub-Saharan Africa, most of the continent's cycad species are relatively localized (Melville, 1957; Goode, 1989, 2001; Jones. 1993; Donaldson, 2003). In South Africa, several species are confined to single isolated inselbergs or outcrops (Goode, 2001 ; Donaldson, 2008). Forty-five taxa occur in a single country only, 18 are found in two countries, and only five occur in three countries or more (Donaldson, 2003). The majority of species are found in southern Africa, and with 38 species, South Africa is a regional centre of cycad diversity (Goode, 1989, 2001; Grobbelaar, 2004; Donaldson, 2008). Along with Australia and Mexico, South Africa is also a global cycad diversity hotspot (Osborne, 1990a; Donaldson, 2003, 2008). Other African countries with comparatively high cycad diversity (5-14 taxa) are Mozambique, Swaziland, Democratic Republic of Congo, Kenya, Tanzania and Uganda (Goode, 1989, 2001; Donaldson, 2003). It may appear that the high diversity in South Africa is due to more fieldwork and taxonomic study having been conducted in this region than in other countries, but since the turn of the century there has been an increase in fieldwork on cycads in other African countries with several new species being described, but this has not changed the overall pattern of species richness (Donaldson, 2003). However, additional fieldwork undertaken in parts of the distribution of Encephalartos in Central Africa where large gaps exist (e.g., Cameroon, Nigeria and Democratic Republic of the Congo) may well result in the description of further new species.

Anatomy and Ecophysiology

Plant Parts

Leaves

All cycads have a basic structure comprising leaves, stems, roots, and cones. Their leaves are pinnately compound, usually sclerophyllous, often bear spines on the leaflet margins, and are typically long-lived (1-3 years, or more) (Chamberlain, 1919; Jones, 1993; Whitelock, 2002; Grobbelaar, 2004). The stomata on cycad leaves are primarily located on the abaxial surface in sunken pits, and the epidermis, which is underlain by one or more layers of thick-walled hypodermis cells, is typically coated with a thick cuticle (Vorster, 2004). The leaves of some Encephalartos species--the so-called "blue leaf' taxa (e.g., E. horridus, E. lehmannii, E. princeps and E. trispinosus), produce a glaucous waxy layer on the leaf surface which assists in reducing evapotranspiration (Grobbelaar, 2004) (Fig. 2d). The petiole of each leaf has a swollen portion at the point of attachment to the stem, which is commonly referred to as the leaf base. This structure persists on the stem after the leaf has died and been abscised, and together with other leaf bases, forms a protective outer stem covering equivalent to bark on dicotyledonous trees (Chamberlain, 1919; Giddy, 1984; Grobbelaar, 2004). Except for Stangeria eriopus, the stem apices of all cycad species bear short, triangular, woody, often tomentose, modified leaves with a pointed free end, known as cataphylls (Fig. 3c) (Grobbelaar, 2004). These structures are formed immediately after the production of new leaves or cones, and provide protection to the stem's sensitive growing point (Grobbelaar, 2004).

Stems

A cycad stem generally consists of a large, starch-rich pith in the centre (with numerous medullary rays, and no growth rings), surrounded by a narrow zone of vascular tissue, followed by a thick layer of cortex, and an outer layer of leaf bases and the remains of dead cataphylls (Chamberlain, 1919; Jones, 1993; Norstog and Nicholls, 1997; Grobbelaar, 2004; Vorster, 2004). In very old stems of some Encephalartos and Cycas species, the leaf bases and cataphylls are shed at the base to make way for a periderm layer (Grobbelaar, 2004). Stangeria eriopus does not retain its leaf bases, its stem being protected by only a thin periderm-derived cork layer (Grobbelaar, 2004). With the exception of some species (e.g., E. transvenosus and E. woodii), cycad stems do not produce axillary buds and therefore seldom form aerial branches, although many species produce subterranean lateral branches referred to as suckers or pups (Grobbelaar, 2004). In some Encephalartos species (e.g., E. villosus, Fig. 2c) and Stangeria eriopus (Fig. 2e), the stem remains underground (subterranean) for the entire lifespan of the plant (Fig. 2e). Other species are semi-subterranean, with only partially emergent stems, (e.g., E. cycadifolius, Fig. 2b), or dwarf arborescent (shrub-like) (e.g., E. horridus, Fig. 2d), while true arborescent species typically attain heights >3 m (e.g., E. latifrons, Fig. 2a, Cycas thouarsii Fig. 2f), with the tallest Encephalartos species CE. transvenosus) reaching up to 14 m (Goode, 1989, 2001; Grobbelaar, 2004).

Cones

Cycads are dioecious, usually with notable differences in the size and shape of the cones between sexes within species (Grobbelaar, 2004). While sex is generally considered fixed in cycads, stressful conditions can occasionally alter the sex of individual plants (Osborne, 1990b; Osborne & Gorelick, 2003, 2007). In Encephalartos and Stangeria, male and female cones (referred to as microsrobili and megastrobili, respectively) typically consist of a central axis to which numerous tightly-packed cone scales (termed microsporophylls in males and megasporophylls in females) are attached (Fig. 4) (Jones, 1993; Grobbelaar, 2004). Each megasporophyll bears two naked seeds (i.e. not enclosed in an ovary), each surrounded by a sarcotesta--a fleshy, often brightly-coloured seed covering. The female cones of Cycas species differ markedly from all other cycad genera and are thus often referred to as pseudocones (Fig. 4h) (Jones, 1993; Grobbelaar, 2004). The megasporophylls are leaf-like and loosely grouped, each typically consisting of a stalk with a comb-like terminal section, or bulla. The ovules are borne on the lateral sides of the stalk with their micropyles directed away from the cone axis. Unlike the females, Cycas male cones are similar in structure to those of other cycad genera (Fig. 4g).

As they mature, the sporophylls of male cycad cones separate from one another, thus exposing the opening pollen sacs on their undersides in order to release the pollen contained therein (Grobbelaar, 2004). The male cones of many cycad species exhibit thermogenesis and odour emission to attract insect pollinators (Tang, 1987; Terry et al., 2004, 2007). Cones are attached to the stem apex via a peduncle, which may or may not be visible. In Encephalartos, some species produce only one cone per plant per coning event (e.g., E. ngoyanus), while others may produce multiple cones, which emerge either simultaneously (e.g., E. friderici-guilielmi) or successively (e.g., E. ferox) (Fig. 4c,d) (Rousseau, 2013). The cones of E. aemulans, E. heenanii, and the seven so-called "fine-leaved, woolly-coned" Encephalartos species (E. brevifo lio latus, E. cycadifolius, E. friderici-guilielmi, E. ghellinckii, E. humilis, E. laevifolius and E. lanatus) endemic to South Africa are covered with a dense layer of persistent, fine, whitish-yellow hairs often referred to as "wool" (Fig. 4a,b) (Goode, 2001; Grobbelaar, 2004). Stangeria eriopus cones also have a woolly covering, but it is less dense, and greyish-brown in colour (Fig. 4e,f). The cones of some species, such as E. senticosus, initially possess a woolly covering, but most of it is lost as the cone matures (Grobbelaar, 2004).

Root Systems

All cycads have two different kinds of root systems, with a third type unique to the genus Cycas (Jones, 1993). The primary root system is a positively gravitropic taproot system, with lateral roots branching off the main taproot (Jones, 1993; Grobbelaar, 2004). In addition to the essential functions of anchorage and uptake of water and nutrients, the taproot is often swollen, and serves as a store of water and carbohydrates (Jones, 1993; Grobbelaar, 2004). The secondary root system comprises highly specialised lateral roots, which are negatively gravitropic and hence grow upwards, forking repeatedly and thereby forming coral-like structures that are termed coralloid roots (Fig. 3a) (Jones, 1993; Grobbelaar, 2004). These roots occur just beneath or at the soil surface and contain symbiotic cyanobacteria (blue-green algae), which fix atmospheric nitrogen (Grobbelaar et al, 1986; Chang et al., 1988; Jones, 1993; Grobbelaar, 2004). This additional source of nitrogen enables the plants to survive in the often nutrient-poor substrates on which they grow (Vorster, 2004). Cycas species produce adventitious roots that arise from the lower side of trunk offsets and grow downwards through the air in close proximity to the trunk, and often also through the cortex (Jones, 1993). Cycad species with subterranean stems (e.g., E. villosus) have contractile roots that periodically contract lengthwise, pulling the stem deeper into the soil to counteract its upward growth (Grobbelaar, 2004).

Growth, Resprouting and Asexual Reproduction

Cycads are known to have very slow growth rates, usually only producing one set of new leaves annually, but intervals between leaf production events can sometimes span several years (Dyer, 1965; Jones, 1993; Whitelock, 2002). Most of the large, relatively fast-growing South African Encephalartos species such as E. altensteinii, E. ferox, E. natalensis and E. transvenosus take 10-15 years to reach coning size when grown from seed under garden conditions (Grobbelaar, 2004). The fastest maturing Encephalartos species in cultivation is possibly Encephalartos cerinus, which can reach coning size within five years from seed (W. Van Eeden. pers. comm.). Cycads exhibit "apical sprouting", i.e. onward growth from the meristematic tissue at the stem apices (Clarke et al., 2013). In addition to apical sprouting, a few cycad species produce adventitious branches aboveground, and many species form subterranean branches (offsets), which are commonly refen'ed to as suckers or pups (Jones, 1993; Grobbelaar, 2004). Since cycads lack axillary buds, these adventitious buds arise from callus or cortex tissue (Jones, 1993). Chamberlain (1919) noted that in cases of severe injury, such as the burning or removal of the upper portion of an Encephalartos stem, one or more buds may develop on the cortex after a long resting period, and form a new branch (or branches) from the original stem. Similarly, Giddy (1984) showed that after the stem apex of an E. humilis individual under cultivation having completely rotted, it produced several new apices (crowns) where the original stem apex had been, as well as numerous basal suckers.

Offsets may be produced naturally as a form of asexual reproduction as the plant matures, or in response to severe damage to large plants (e.g., when the stem apex is destroyed). Giddy (1984) noted that basal suckering also occurs in tall, single-stemmed species as the stem begins to recline due to its sheer weight. The subterranean caudices of S. eriopus are able to regenerate and produce new leaves after being severed or partially harvested by medicinal plant gatherers (Douwes et al., 2004). Suckers are often removed by cycad growers for asexual propagation once they reach a critical size; however, if they are left to continue growing, the parent plant eventually becomes multi-stemmed. Some species such as E. cupidus and E. cycadifolius produce suckers prolifically, while subterranean-stemmed species such as E. villosus, E. umbeluziensis and E. ngoyanus rarely produce suckers (although sucker production in E. villosus under cultivation can be substantial) (Grobbelaar, 2004). Donaldson (1995a) noted that in E. cycadifolius, the suckers regularly produced by adult individuals may develop their own root systems so that established adults can potentially persist indefinitely. In a study of several large E. cycadifolius populations, no adult mortality was recorded over five years, but individual stems regularly died (Donaldson, 1995a). A much less common phenomenon is the production of aerial suckers (well above ground level), which is observed in E. laurentiamis, often as a result of damage to the main stem (Goode, 2001).

Light, Temperature and Soil Requirements

African cycads occupy a diverse array of habitats, but are mostly found in grasslands, forests and savannas (Goode, 1989, 2001; Grobbelaar, 2004; Donaldson, 2008). Vorster (2004) suggests that cycads often occupy relatively harsh habitats (e.g., nutrient-poor soils in dry areas) in order to avoid competition with faster-growing Angiosperms. Most Encephalartos species grow in sunny positions, but some occur almost exclusively in forest understories or in dense bush, e.g., E. aplanatus and E. villosus (Goode, 2001 ; Grobbelaar, 2004). Savanna species appear to tolerate a range of conditions from open habitats with minimal tree cover to closed-canopy forest-like habitats (Donaldson, 2008). Stangeria eriopus occupies a very broad habitat niche, from open grasslands in which it typically produces short leaves (25 cm long), to forests, where its leaves reach up to 2 m in length (Goode, 2001) Seedlings of most Encephalartos species generally cannot cope with direct sunlight: the first leaf produced soon senesces, and the plant subsequently produces progressively smaller leaves until it dies (Grobbelaar, 2004). For seedlings of species that occur in sunny habitats to establish, seeds must germinate in microhabitats that provide shade for several years (Grobbelaar, 2003, 2004). Suitable microhabitats commonly occur in rock crevices or among scattered boulders, but tall grass or shrubs can also provide the necessary initial shady environment (Grobbelaar, 2004).

Most Encephalartos species occur in frost-free regions, however, the seven "fineleaved, woolly-coned" South African species are tolerant of sub-zero temperatures, and are found mostly in high-lying areas that receive frost and sometimes snow in winter (Goode, 2001; Vorster, 2004). Encephalartos species frequently occur between rocks and boulders on steep mountain slopes and in ravines, and hence grow in very well-drained substrates (Jones, 1993; Goode, 2001; Grobbelaar, 2004). Occupying these relatively harsh microsites necessitates the storage of water in stems and swollen roots (Vorster, 2004). In cultivation, Encephalartos species generally perform poorly if planted in heavy clay soils that become waterlogged, as their water storage organs cannot cope with inundation (Grobbelaar, 2004; Vorster, 2004). They do, however, grow successfully in soil that is well-drained and enriched with organic matter, often growing much faster and more luxuriantly than in their natural habitats (Vorster, 2004; Grobbelaar, 2004). African cycads (and cycads in general) occur in a wide range of soil types, with only a few species having fairly specific soil requirements, such as the South African species E. dolomiticus and E. inopinus, which are restricted to calcareous, slightly alkaline soil of dolomitic origin (Grobbelaar, 2004).

Role in the Ecosystem

The role that cycads play in their associated ecosystems is generally poorly understood (Donaldson, 2008). While it is well-established that all cycads produce coralloid roots which host symbiotic nitrogen-fixing cyanobacteria, the impact of this form of nitrogen fixation on nutrient dynamics is unknown (Grobbelaar, 2004; Donaldson, 2008). Although cycads are mostly avoided by herbivores, there are various insects that use them as a food source, some of which are dependent on their cycad host for survival (Vorster, 2004). In South African Encephalartos there may be up to 12 cycad-specific insects occurring on a single species (Donaldson, 2008). Insects associated with cycads may feed on the leaves, cone material (including pollen) or seeds (e.g., Antliarhinus zamiae, Fig. 5c). (For interactions between insects and cycad cones, see subsequent section on pollination ecology.) In the case of weevils associated with cycads, these insects are often dependent on the plants in such a tightly linked symbiotic relationship, that extinction of the cycads would result in extinction of the insects as well (Oberprieler, 1995a). Furthermore, when cycad populations are in cone, they can produce substantial resources for local wildlife (see subsequent section on seed dispersal), but the extent to which these animals depend on these resources is unknown (Donaldson, 2008).

Relatively few herbivores (both vertebrates and invertebrates) feed on the sclerophyllous, often spiny, and generally toxic leaves of cycads, since they are difficult to chew and digest, and in the case of livestock, can even be fatal (Schneider et al., 2002; Grobbelaar, 2004; Bayliss et al., 2009; Prado, 2011). Zanzibar red colobus monkeys feed on the leaves of the arborescent cycad Encephalartos hildebrandtii, showing a preference for young, tender leaves over older ones (Nowak & Lee, 2011). The monkeys have developed specialised feeding techniques that enable them to reach the newly-emerging soft leaves, which contain less of the toxin hydrogen cyanide than older leaves (Nowak & Lee, 2011). The monkeys probably tolerate the toxins by detoxifying them in the gut, avoiding lethal overdoses, and consuming charcoal, which may aid in toxin adsorption (Nowak & Lee, 2011). Encephalartos stems are known to be utilized as a source of food and/or water by porcupines (Goode, 1989). Singh (2012) indicates that porcupines tend to feed on the stem material near the base of the trunk, making it vulnerable to being toppled over by large mammals. It is not known to what extent damage caused by porcupines results in the mortality of adult plants.

Insects that feed on cycad leaves are mostly Lepidopteran larvae--both butterflies and moths--which are resistant to the toxins, and often sequester and incorporate them into their own survival strategies (Donaldson & Bosenberg, 1995; Staude, 2001; Vorster, 2004; Bayliss et al., 2009). However, there are no records of butterfly species feeding on African cycads, and the vast majority of the continent's cycad-feeding moths (affecting both Encephalartos and Stangeria) belong to the subfamily Diptychinae in the family Geometridae (Staude, 2001). Evidence suggests that all moth species in the Diptychinae rely on cycads as a food source at some point during their larval stage (Staude, 2001). Staude (2001) listed 16 moth taxa from six different genera as being recorded feeding on the leaves of African cycad taxa, the largest genus being Callioratis (five species) (see Callioratus abraxus in Fig. 5d). Bayliss et al. (2009) noted a sixth undescribed Callioratis species, and documented a seventh (C. grandis) feeding exclusively on Encephalartos gratus in Malawi, bringing the total number of known African cycad leaf-feeding moth taxa to 18. C. grandis was shown to favour mostly large is. gratus plants, and out of the 532 cycads inspected, 270 (51%) presented evidence of moth damage (Bayliss et al., 2009). Despite the damage sometimes being extensive, it did not appear to be fatal, with cycads producing new leaf flushes during the growing season following the damage (Bayliss et al., 2009).

In warm, humid parts of southern Africa, especially South Africa's KwaZulu-Natal province, the leopard magpie moth (Zerenopis lepida) is a common cycad pest in gardens (Fig. 5a,b) (Donaldson & Bosenberg, 1995; Staude, 2001; Grobbelaar, 2004). The female moth lays her eggs on tender, newly-emergent cycad leaves, and the larvae consume all the soft parts of the leaves within a few days (Grobbelaar, 2004). In the wild, a large percentage of Z. lepida larvae leave their cycad host after their third moult to feed on a suite of secondary larval food plants (Staude. 2001). This phenomenon likely ensures that their often limited primary food source is not permanently damaged (Staude, 2001). Donaldson & Bosenberg (1995) showed that the cycad toxin macrozamin is an essential component of the diet of early instar Z. lepida larvae, but after the third moult it is no longer required. Stangeria eriopus (cultivated and wild), Cycas thouarsii (cultivated), and 20 Encephalartos species (19 wild and one cultivated) have been recorded as hosts for at least one Z. lepida life history stage (Donaldson & Bosenberg, 1995). Z. lepida also occasionally destroys the seeds of Encephalartos species by feeding opportunistically on the gametophyte (Donaldson, 1993).

Fire Ecology

There are very few scientific studies on the fire ecology of African cycads (e.g., Donaldson, 1995a), but popular articles, species descriptions and accompanying photographs in books (e.g., Goode, 2001), as well as field observations (e.g., Melville, 1957) and anecdotes in broad-based cycad publications (e.g., Dyer, 1965) indicate that many Encephalartos species are fire-tolerant and some are fire-stimulated. Dyer (1965) purports that Stangeria eriopus is also responsive to the stimulus of fire, but this is not confirmed elsewhere in the literature. Donaldson (1995b) suggests that fire-stimulated leaf and cone production in cycads is possibly linked to the overall life history of the species in question, and is more prevalent amongst taxa that have persistent adults that tend to produce large numbers of suckers. In some Encephalartos species, the leaves may be completely incinerated by fire, but the plants subsequently recover by producing new leaves from the stem apices (Goode, 2001). The tightly-packed, persistent leaf bases covering Encephalartos stems to a thickness of approximately 2.5-7.5 cm appear to provide effective insulation against the heat from fires (Dyer, 1965). However, while adult plants of most Encephalartos species appear to be fire-tolerant, seedlings are reportedly very sensitive, and those that recruit in protected rocky sites usually have the highest survival rates (Melville, 1957; Rousseau & Rousseau, 2011; Cilliers, 2012). The occurrence of too-frequent fires has been invoked as a contributing factor to the poor recruitment observed in wild populations of E. heenanii (Hurter, 1994) and the failed establishment of reintroduced seedlings of E. middelburgensis (Harris & Harris, 2003).

All the fine-leaved, woolly-coned South African Encephalartos species (E. brevifoliolatus, E. cycadifolius, E. friderici-guilielmi, E. ghellinckii, E. humilis, E. laevifolius and E. lanatus) grow in grasslands that experience recurrent natural fires (Goode, 2001). These species are often observed with blackened stems post-fire in the wild, and are therefore considered fire-tolerant, with strong evidence for some being fire-stimulated. Many of the broad-leaved Encephalartos species survive periodic fires in savanna habitats, while some taxa occur in forests where fires are highly unlikely (Goode, 2001). Dyer (1965) quoted a farmer near Bedford in the Eastern Cape, South Africa, who stated that the year after burning part of an E. cycadifolius population on her farm in late winter, the plants produced new leaves, and there was "scarcely one in hundreds that had not fruited". By contrast, plants in other parts of the population that did not burn showed no signs of coning. Donaldson (1995a) confirms the stimulatoiy effect of fire on both the production of new leaves and cones in E. cycadifolius, with strong evidence for fire-stimulated mast coning at firebreaks, where all the mature plants on the burned side produce cones, whereas those on the unburned side do not. Raimondo and Donaldson (2003) note that these coning events typically occur synchronously two years post-fire.

It also appears that fire stimulates vigorous leaf and/or cone production in E. ghellinckii, E. humilis, E. lanatus and E. laevifolius. Giddy (1984) suggested that the almost annual production of new leaves and cones in E. lanatus populations was due to the regular fires they were subjected to. Experimental burns conducted on a large collection of E. humilis (>30 individuals) under cultivation at the Lowveld National Botanical Garden in Nelspruit, South Africa, resulted in a substantially higher proportion of burned individuals producing new leaves in the growing season following the winter burn than did adjacent unburned plants (Van der Walt, 2010). Cones were also produced by some individuals that had only coned once in the 15 years prior to the fire (Van der Walt, 2010). The necessity for fire to stimulate leaf and cone production in E. humilis is also suggested by the generally poor performance of plants in cultivation that are unburned compared to the lush growth and higher coning frequency of wild plants (Goode, 2001). Similarly, observations of a wild population of E. ghellinckii, half of which was exposed to fire, revealed that the burned individuals showed nearly 100% cone production, while no cones were produced by the unburned plants (Zunckel, 1990). This species' apparent dependence on fire for reproduction was also reflected in a demographic study of six E. ghellinckii populations by Scott-Shaw (1995), which showed that the populations from which fire was excluded for seven years or more showed the lowest coning and seedling recruitment rates. Observations by Zunckel (1995) pointed towards possible fire-stimulated leaf and cone production in E. laevifolius, although it did not appear that populations were dependent on fire for survival and seedling recruitment.

Hence, while it appears that most African cycad species are fire-tolerant, and that the fine-leaved, woolly-coned Encephalartos species generally show the most pronounced positive response to fire, the phenomenon of fire-stimulated leaf and cone production is still poorly understood, and requires further study. Further investigation is needed to determine the actual fire-related cues that trigger leaf and cone production: heat pulse, smoke, altered post-fire environment (presence of ash, release of nutrients, increased light and space availability, etc.), or a combination thereof. Knowledge on the most appropriate burn frequency and season would also be useful for providing ecologicallysound fire management guidelines for cycads in fire-prone habitats.

Population Ecology

Life History Patterns and Population Size Structure

While many studies on cycad population ecology have been conducted in Australia, Asia and Central and South America, similar published studies on African cycads are limited. From 1984 until the late 1990s, the nature conservation authority of the former Transvaal province of South Africa conducted long-tenn population monitoring of the ca. 18 Encephalartos species in the region (Fourie, 1995), but most of these data are unpublished. The African species studied in greatest detail are E. cycadifolius and E. villosus (Donaldson. 1995a,b; Raimondo & Donaldson, 2003), which represent the two extremes in the classification of Encephalartos life histories by Donaldson (1995b). This classification framework divides the genus Encephalartos (with a focus on South African species) into four different life history types (Persister, Persister/ reproducer, Reproducer/persister and Reproducer), each with a distinctive growth fonn and number of cones produced during a single coning event. Each category falls along a continuum from one extreme at which clonal reproduction and persistence are dominant traits of the life history (Persister), to the other extreme, where sexual reproduction takes precedence (Reproducer).

Persisters are long-lived (hundreds of years), tend to produce large numbers of basal suckers, with a single cone per stem (at least for females), and exhibit infrequent sexual reproduction (Donaldson, 1995b). This life history type is generally associated with arid environments or grasslands, and shows a tendency to mast seed in response to cues such as fire or rainfall (e.g., E. cycadifolius, E. humilis and E. trispinosus) (Donaldson, 1995b). Consequently populations of these taxa typically display J-shaped size class distributions characterised by very few seedlings and a preponderance of adults (Raimondo & Donaldson, 2003). At the opposite end of the spectrum, Reproducers are relatively short-lived (decades), they have single (often subterranean) stems with no basal suckers under natural conditions, produce only one cone per plant (at least for females), and rely heavily on sexual reproduction (e.g., E. caffer, E. umbeluziensis and E. villosus) (Donaldson, 1995b). Species in this group either inhabit grasslands or forest margins and/or understories, and their population size class distributions are typically reverse J-shaped, with a high number of seedlings and few mature plants (Raimondo & Donaldson, 2003).

The two life history types in the middle of the continuum exhibit a more complex combination of persistence and reproductive traits, but one or the other is more evident (Donaldson, 1995b). Persister/reproducers possess aerial stems several metres in height, with several basal suckers, and produce a single (often large) cone per plant per coning event (females only) (e.g., E. longifolius and E. lehmannii). Taxa of this life history type generally occur in xeric environments or fire climax vegetation, exhibiting a high incidence of mast seeding in the latter (Donaldson, 1995b). Reproducer/persisters have a similar growth form to persister/reproducers, but sucker less and produce several cones per plant during a given coning event (e.g., E altensteinii, E. natalensis and E. transvenosus). They occur in woodlands (moist savannas), forests and grasslands, displaying a high incidence of mast seeding in grasslands, but not in forests or woodlands. Published size class distribution data for species in these two groups are scant, but they possibly exhibit population structures marked by more even representation of the various size classes with spikes in the seedling and juvenile classes (e.g., E. longifolius in Singh (2012)) following masting events.

Population Size

Several African cycad species already had very restricted distributions and occurred in small numbers before the advent of intensive cycad collecting (e.g., E. brevifoliolatus, E. dolomiticus, E. dyerianus, E. latifrons, E. nubimontanus and E. tegulaneus subsp. powysii) (Donaldson, 2003). It has often been these naturally rare taxa that have been targeted most heavily by cycad collectors, with E. brevifoliolatus and E. nubimontanus now both Extinct in the Wild (EW) as a result (Donaldson, 2008). The 2003 IUCN Red List assessment for African cycads indicates that all 18 Critically Endangered (CR) Encephalartos species had population sizes of <950 plants, and six had estimates of <100 plants remaining in the wild (Donaldson, 2003). Other taxa have much broader distributions and larger populations, numbering in the tens of thousands (e.g., E. villosus, with an estimated 100,000 plants in the wild in 2003) (Donaldson, 2003). Within species, population size may also vary by order of magnitude, e.g., E. ghellinckii, with populations ranging from just eight individuals to several hundred (Scott-Shaw, 1995), and E. laevifolius, which in 1990 had seven known populations ranging from 25 to [+ or -]500 individuals (Zunckel. 1990).

Donaldson (1995b) notes that population size does not necessarily mean the same thing for different cycad species. For example, a population of 500 long-lived adult E. cycadifolius individuals (persister type life history) may be far less threatened than an E. villosus population consisting of 1000 relatively short-lived adults with a reproducer type life history (Donaldson, 1995b). If the E. villosus population undergoes reproductive failure (e.g., due to pollinator limitation) it has no other means of perpetuating itself (e.g., by asexual reproduction), whereas the E. cycadifolius population can persist in the absence of pollinators by continued basal suckering. Notwithstanding, Raimondo & Donaldson (2003) showed using matrix modelling that population growth in both E. cycadifolius and E. villosus (and, by extension, all cycads) is sensitive to changes in the abundance of adult plants, with rapid declines occurring under adult plant harvesting scenarios similar to those experienced by species popular in trade. For E. cycadifolius, even a 5% adult harvest on an annual basis caused the population to exhibit negative growth (Raimondo and Donaldson, 2003). Despite similar responses to adult mortality, the two species displayed markedly different recovery times, with E. villosus recovering from a major loss of adults in <25 years given no other perturbations, whereas E. cycadifolius may take 100-300 years to recover from a similar harvesting event. Raimondo & Donaldson (2003) therefore recommend that seeds be harvested from wild populations instead of adult plants, as models suggest that even high levels of seed harvesting have little impact, although this would require careful monitoring.

Reproductive Ecology

Pollination Ecology

Following an influential publication by Chamberlain (1935), the prevailing cycad pollination paradigm was one of wind-pollination. This assumption was accepted despite earlier observations on African cycads that suggested that insects played an important role in cycad pollination (Pearson, 1905; Rattray, 1913; Marloth, 1914). A growing body of experimental evidence in recent years corroborates the deductions made from these early observations, and entomophily is now considered widespread in cycads (Grobbelaar, 2004). Beetles (Coleoptera) have been identified as the predominant pollinators, with most pollinator species belonging to the superfamilies Curculionoidea and Cucujoidea (Obcrprieler, 1995a, b, 2004). Some insects visit cycad cones to lay their eggs in the sporophylls or central cone axis for their larvae to feed on, while others feed on the pollen in the cone, or carry it to their nests to feed their larvae (e.g., honey bees, Apis melifera) (Grobbelaar, 2004). Not all insects that visit the male cones visit the female cones as well, and therefore only a proportion of them play a role in pollination (Grobbelaar, 2004). Among the insect species that are pollinators of Encephalartos, there also appear to be both generalists (pollinate several species) and specialists (pollinate a single species only) (Donaldson, 2004a). Geographical distribution patterns of insects associated with cycads across Africa show apparently more diverse insect assemblages (with a variety of potential pollinators) in the south, with less diversity in the north (Donaldson, 2004a).

There is an emerging body of literature on odour emission (volatiles) and thermogenesis (heat production) in cycad cones, which shows that the volatiles are important for attracting insect pollinators to the cones, while thermogenesis in male cones appears to enhance the volatilization of odours, promote cone elongation and separation of the sporophylls, and increase pollinator activity, growth and probability of survival (Tang, 1987, 1993; Tang et al., 1987; Terry et al., 2004; Suinyuy et al., 2009, 2010, 2012; 2013a). Jacot-Guillannod (1958) documented thermogenesis in the male cones of E. altensteinii and E. lehmannii, and thermogenesis was also studied in male and/or female cones of E. altensteinii, E. barteri, E. bubalinus, E. ferox, E. gratus, E. hildebrandtii, E. longifolius, E. manikensis, E. villosus and Stangeria eriopus (Tang, 1987; Tang et al, 1987). A study on the scent chemistry and thermogenesis in the cones of E. natalensis by Suinyuy et al. (2010) showed that volatile emissions and thermogenesis occurred in association with insect activity on the cones, suggesting that they both play a role in regulating insect behaviour. Three beetle species were found in male E. natalensis cones during pollen shed, two of which were also observed on receptive female cones, and were mainly active between the tightly-packed megasporophylls, strongly suggesting their role in pollination (Suinyuy et al., 2010) (Table 1). A study by Suinyuy et al. (2013b) on variation in the chemical composition of the cone volatiles of 19 Encephalartos species revealed that a total of 152 compounds were identified among the 19 species, the most common of which were monoterpenes, nitrogen-containing compounds and unsaturated hydrocarbons. It was also shown that male and female cones emitted similar volatile compounds, and volatile variation in the genus likely reflects both phylogeny and adaptations to specific beetle pollinators (Suinyuy et al., 2013b).

To date, detailed experimental pollination studies have been conducted on five African cycad species: E. cycadifolius (Donaldson et al., 1995), E. villosus (Donaldson, 1997), E. transvenosus (Grobbelaar, 1999), E. friderici-guilielmi (Suinyuy et al., 2009) and S. eriopus (Prochec & Johnson, 2009), all of which documented pollination by beetles, except for the study on E. transvenosus (Table 1). In this species, no insects were observed in female cones, but male cones were always infested with the pollen-feeding weevil Metacucujus encephalarti (Grobbelaar, 1999). Pollinator exclusion experiments also showed that insects do not appear to be involved in the pollination of this species, and wind may therefore play a role (Grobbelaar, 1999). Stobart (1989) found nine beetle species on E. altensteinii cones, four of which were regularly observed on the cones of both sexes, and evidence of pollen transfer from male to female cones was found for two species. Zunckel (1995) observed a weevil in the genus Porthetes on the male cones of E. laevifolius, but did not find the same insect in the female cones. Both Zunckel (1995) and Grobbelaar (2004) noted that in E. laevifolius and all other woolly-coned Encephalartos species, the dense covering of hairs on the outside of the cone makes it very difficult for wind to transport pollen to the inside, and that pollen-laden insects would be required to push through the woolly layer to reach the seeds and effect pollination.

Donaldson et al. (1995) showed that pollen-laden weevils coated in dye and released near female cones of E. cycadifolius entered the cones, and the dye (and hence the pollen) was subsequently detected on the micropylar ends of the ovules. The study demonstrated that two species of cucujoid beetles (Coleoptera: Cucujoidea): Metacucujus encephalarti (Boganiidae) and an undescribed Erotylidae species were the principle pollinators of the species. For E. villosus, pollinator exclusion experiments revealed that when insects were excluded from female cones by either net bags or insecticide, the proportion of fertilized ovules significantly decreased (Donaldson, 1997). Out of five beetle species that were found on E. villosus cones at the time of pollination, an undescribed weevil (Porthetes sp., Curculionidae) was consistently the most important pollinator. Antliarhinus zamiae and an undescribed beetle species in the Xenoscelinae (Languriidae) played a minor role in pollination, and Metacucujus goodei (Boganiidae) and a second species of Xenoscelinae appeared to have little or no effect (Donaldson, 1997). Although wind exclusion from female E. villosus cones resulted in a mean decrease of 14.5% in seed set compared to naturally pollinated cones, Donaldson (1997) argued that the influence of wind may have been an artefact of the exclusion methods. Furthermore, given that E. villosus is an understorey plant with low coning frequencies and large distances between reproductive males and females (up to 1 km), it appears that wind plays a negligible role in the pollination of the species (Donaldson, 1997).

In a pollination study on E. friderici-guilielmi, Suinyuy et al. (2009) showed that two beetle species, Porthetes hispidus (Curculionidae) and Metacucujus encephalarti (Cucujoidea) were capable of transferring pollen to the micropyles of receptive ovules. Furthermore, enclosure experiments demonstrated that batches of P. hispidus, M. encephalarti and an undescribed Erotylidae species (Cucujoidea) enclosed individually in mesh bags over female E. friderici-guilielmi cones resulted in seed set that was similar to that of open control cones, and significantly higher than that of cones from which all insects were excluded. These findings therefore confirm insect pollination in E. friderici-guilielmi (Suinyuy et al., 2009). Suinyuy et al. (2010) demonstrated that insect activity in male cones of E. natalensis occurred simultaneously with volatile emission, thermogenesis and pollen shedding, which suggests that volatiles and thermogenesis both play a role in regulating insect behaviour. Three beetle species, all of which belonged to genera involved in the pollination of other Encephalartos species, were active and covered in pollen in male E. natalensis cones (Suinyuy et al, 2010). The three beetle species were also present on female cones, which did not exhibit thennogenesis, but did emit volatiles. Circumstantial evidence from this study therefore strongly suggests that E. natalensis is also insect-pollinated.

Prochec & Johnson (2009) studied the pollination biology of S. eriopus, and showed that in contrast to most other cycads, this species showed no evidence of cone thermogenesis. A striking finding was that the odour of both male and female Stangeria cones mimics that of fennented fruit (Proches & Johnson, 2009). It is therefore possible that the pollination system of Stangeria is deceptive in nature, involving mimicry of angiosperm fruits (Proche? & Johnson, 2009). A large array of insect visitors were observed on the cones, the most common being sap and rove beetles (Coleoptera: Nitidulidae, Staphylinidae), and fruit flies (Diptera: Drosophilidae). Of these, only the sap beetles were able to effectively pollinate S. eriopus under experimental conditions.

Coning Patterns and Sex Ratios

Reproduction in most South African (and probably most African) Encephalartos species is infrequent and irregular (Donaldson. 2008). Synchronous reproduction (mast seeding) occurs to some degree in most species, resulting in distinctive 'coning years' in which most adult plants in a population produce cones followed by an interval in which few or no seeds are set (Donaldson, 1993). In order to test the 'predator satiation hypothesis' in cycads, Donaldson (1993) studied masting and seed predation in eight South African Encephalartos species (18 populations altogether). The predator satiation hypothesis proposes that the superabundance of seeds during mast years will significantly outweigh the capacity of seed predators to consume them (Silvertown, 1980). The primary seed predators of Encephalartos seeds were Antliarhinus zamiae (Fig. 4c) and A. signatus - two weevils that develop exclusively on cycad seeds (Donaldson, 1993). Masting intensity was greatest in populations that experienced the highest levels of pre-dispersal seed predation in years when few individuals coned. Conversely, the lowest masting intensity was observed in populations with minimal levels of seed predation and few or no weevils present (Donaldson, 1993). Larger seed crops appeared to result in lower seed predation levels by A. signatus and A. zamiae in four E. altensteinii populations. Furthermore, in populations of five cycad species, differences in seed crop size accounted for 48-66% of variation in seed predation levels (Donaldson, 1993). These results are congruent with the predator satiation hypothesis. Notwithstanding, in most Encephalartos populations, there did not appear to be a significant reduction in seed predator numbers after 2-8 years between masting events. Moreover, in two of three populations examined, periodic reproduction did not increase the proportion of seeds surviving to dispersal over a four year period. These results were interpreted to mean that periodic reproduction has not resulted from selection imposed by seed predators, but that selection may favour individuals that experience lower levels of seed predation by coning synchronously with most of the reproductive plants in the population (Donaldson, 1993).

Comprehensive studies of large Encephalartos populations indicate that the sex ratio of populations is typically 1:1 (Donaldson, 2008). This scenario has not been documented universally, however. Out of the 124 reproductive Encephalartos gratus individuals examined by Bayliss et al. (2009), 77% were male and 23% female. Prochec & Johnson (2009) showed that the sex ratio (male: female) in two S. eriopus populations was 3:1 and 2:1. Males also produced more cones than females, resulting in an average of 3.5 and 2.0 male cones per female cone at the two populations, respectively (Prochec & Johnson, 2009). Grobbelaar et al. (1989) determined the sex ratio of 638 reproductively mature cycads in an E. transvenosus population in Limpopo Province, South Africa. The overall ratio of males to females was 2.3:1, while that of plants in the taller size classes (2.6-9.5 m) was 7:1. This highly skewed sex ratio in larger (and hence older) plants was thought to be as a result of higher mortality of old female plants and/or a more significant decline in cone production by ageing females compared to males (Grobbelaar et al, 1989). Several possible reasons were cited for the male-biased sex ratio observed throughout the population, one of which was that male plants may have shorter intervals between coning events, since their cones are much less energetically expensive to produce than female cones. Hence, the true sex ratio may have been closer to 1:1. but the higher coning frequency in males relative to females meant that it always appeared that males were in the majority.

Sex ratios also appear to change as cycad populations decline. Many large, vigorous Encephalartos populations have a sex ratio approaching 1:1 (Grobbelaar 1999; Raimondo, 2001), whereas small populations appear to be dominated by males, (e.g., the male: female ratio for E. latifrons is 4:1, the few specimens of E. brevifoliolatus found before the species became extinct in the wild were all male, and the only known original E. woodii individual was also male). While the causal mechanism(s) may include higher mortality of older females relative to males as suggested by Grobbelaar et al. (1989), strongly male-biased ratios in small cycad populations are often due to selective harvesting of females by cycad collectors (Donaldson, 2008). Female plants are more desirable than males, since they produce seed which can be used to propagate and sell plants, whereas males only produce pollen.

Seed Ecology

Seed Set

The maximum number of seeds produced per cone within southern African Encephalartos species (data are not available for all African species) tends to be lowest in the subterranean- and semi-subterranean-stemmed taxa, and highest in the large arborescent species. The diminutive semi-subterranean-stemmed E. humilis exhibits the lowest seed output per cone of all southern African species (maximum of [+ or -]100 seeds/ cone), followed by E. ngoyanus and E. cupidus (110 and 120 seeds/cone, respectively) (Grobbelaar, 2004). The remaining taxa in this group (E. aplanatus, E. coffer, E. cerinus, E. cycadifolius and E. umbeluziensis) produce up to 200-300 seeds/cone, with the exception of E. villosus, in which seed production may reach 460 seeds/cone (Grobbelaar, 2004). Stangeria eriopus is comparable to E. humilis, producing only 35-100 seeds/cone. Some of the tall-growing arborescent Encephalartos species display fairly low seed/cone outputs (<300) relative to their size (e.g., E. dyerianus and E. middelburgensis, which can reach heights of 6 m and 7 m, respectively), but most species of this growth form yield >400 seeds/cone, with an upper bound of almost 800 in E. latifrons and E. transvenosus (Grobbelaar, 2004). Total seed production per plant may exceed these figures, as many species are able to produce more than one female cone per crown simultaneously, and a single plant may have several crowns capable of producing cones. A female E. middelburgensis monitored by Robbertse et al. (1999), for example, produced five cones during a single coning episode, the number of seeds per cone ranging from 258 to 304, with a total of 1408 seeds produced by the plant as a whole.

Viable seed production in Encephalartos species is dependent on population size: in very small populations of <50 individuals, seed set is poor or nil, suggesting an Allee effect, while in populations of >250 individuals, seed set is typically >70% (Donaldson, 2008). Production of only small quantities of fertile seeds appears to be a common phenomenon in small cycad populations in southern Africa and elsewhere (Vovides et al, 1997; Donaldson, 1999; Heibloem, 1999). Seed predation occurs frequently in populations of Encephalartos species, where weevils in the genus Antliarhinus can destroy up to 90% of the seed crop (Donaldson, 1993). In most populations, seeds experience further mortality due to desiccation, resulting in only a small percentage that germinates (Donaldson, 2008). It also appears that viable seed set within populations may vary with individual cone size and environmental factors. Phelan et al. (1993) showed that larger E. ferox cones produced significantly lower percentages of viable seed than smaller ones, and that cones sampled from plants growing in shady conditions contained significantly fewer seeds than those from individuals growing in full sun, but the proportion of viable seeds was significantly lower in the latter.

A major influence on seed set in many cycad populations is pollen limitation due to large distances between male and female plants (frequently exacerbated by gaps created in populations due to cycad collecting, e.g., E. latifrons) as well as skewed sex ratios (often large male to female ratios due to selective removal of female plants by collectors) (Donaldson, 2008). Additionally, intensive collecting of cycads can cause concurrent losses of pollinators to such an extent that the pollinators undergo local extinction (as appears to be the case in E. latifrons (Donaldson, 2004a; Daly et al., 2006)). Several of the known or suspected pollinators of Encephalartos species are associated with only one or a few cycad species (Oberprieler 1995a, b). Surveys of insects associated with cycads in Kenya, South Africa, Zanzibar and Zimbabwe revealed that pollinators are extinct in some cycad populations, which display markedly low concomitant seed production (Donaldson, 1999).

Seed Dispersal

Seed dispersal is one of the most poorly studied aspects of Encephalartos ecology (Donaldson, 2008), although insights are provided by several anecdotal observations of dispersal agents--both biotic and abiotic. Gravity is an important example of the latter, although it usually only results in short-distance dispersal of cycad seeds in close proximity to the mother plant (Fig. 3b) (Grobbelaar, 2004). However, cycads often grow on steep mountain slopes, and upon dehiscence of the cone, the round seeds may roll downslope by gravity, resulting in dense stands of cycad seedlings at the foot of the mountain (Grobbelaar, 2004). Water is also an important dispersal agent: episodic streams that form during heavy rainfall events can disperse cycad seeds over fairly long distances. In Cycas thouarsii, the seeds contain an air-filled spongy tissue (aerenchyma) that aids their flotation on water (Grobbelaar, 2004). Cycas thouarsii is a predominantly coastal species, and its broad distribution, especially on islands, is often ascribed to the dispersal of its seeds by ocean currents (Goode, 2001 ; Grobbelaar, 2004).

Upon disintegration of Encephalartos cones, the generally brightly-coloured sarcotestae that surround the nutrient-rich seeds become visible, and various animals are known to consume either the carbohydrate-rich sarcotesta or occasionally the gametophytes, which have a high starch and protein content (Grobbelaar, 2004; Donaldson, 2008). Observations have been made of consumption and/or dispersal of African cycad seeds by baboons (Papio ursinus), monkeys, elephants (Loxodonta africana), birds (e.g., hornbills), fruit-eating bats, bush pigs (Potamochoerus porcus) and rodents (e.g., vlei rats (Otomys spp.)) (Melville, 1957; Giddy, 1984; Grobbelaar et al., 1989; Grobbelaar, 2004; Donaldson, 2008). Although cycad seeds are usually dispersed over relatively short distances, they may be dispersed over several metres by rodents, which accumulate large reserves in their nests or in hidden caches (e.g., E cycadifolius) (Grobbelaar, 2004; Donaldson, 2008). Some seeds escape damage by the rodents, but slight damage may even be beneficial, as it enhances water uptake and hence aids germination (Grobbelaar, 2004).

Birds such as the crowned hornbill (Lophoceros alboterminatus), trumpeter hornbill (Bycanistes buccinator) and the Cape parrot (Poicephalus robustus suahelicus) are known to carry cycad seeds over long distances to their nests where they consume the sarcotesta and drop the intact kernel (Grobbelaar, 2004). Rock hyraxes (Procavia capensis) and certain fruit-eating bats are also purported to feed on the sarcotesta and disperse the undamaged seeds in the process (Giddy, 1984; Grobbelaar, 2004). Elephants reportedly consume whole female E. poggei cones; the seeds pass through the intestinal tracts unharmed and often germinate in the elephants' dung (Grobbelaar, 2004). Baboons may also disperse seeds over fairly long distances by breaking off cones and using them as 'toys' (e.g., E. septentrionalis, Heenan, 1977) or transporting them to appropriate feeding sites such as rock ledges where the seeds often fall into cracks which offer favourable microhabitats for germination (e.g., E. lanatus) (Grobbelaar, 2004). However, baboons may also be damaging to cycad populations as they frequently break immature whole cones apart, and are known to pull seedlings out of the ground to feed on the remains of the seed (Melville, 1957).

Seed Dormancy, Longevity and Germination

There are no known published studies on the dormancy, longevity and germination of any African cycad taxa in the wild. The available information is based on the propagation of cycads under nursery and garden conditions. Female Encephalartos cones generally disintegrate spontaneously by abscising the sporophylls (cone scales), thereby releasing the seeds, which are usually still immature (Grobbelaar, 1989, 2004). Some species, chiefly four South African arborescent species (E. dolomiticus, E. dyerianus, E. eugene-maraisii and E. middelburgensis) and the fine-leaved, woolly-coned species E. cycadifolius and E. lanatus exhibit a form of serotiny (canopy seed storage) in that the cones do not disintegrate, but gradually dry out, retaining the partially exposed abscised seeds within the desiccated cone (Grobbelaar, 1989, 2004). It is not clear, however, how long the cones stay on the plant after desiccation, but these species probably rely less on gravity and more on biological dispersal agents (e.g., baboons) to free the seeds from the cones. E. transvenosus is unusual in that it retains its seeds in the cones until they are fully developed and therefore able to germinate immediately upon abscission (Grobbelaar, 2004). In some cases E. transvenosus seeds exhibit vivipary, i.e. germination while still attached to, and enclosed in the cone (Grobbelaar, 2004). Since Encephalartos seeds do not enlarge noticeably post-fertilization, it is not possible to distinguish between fertilized (and therefore viable) seed and unfertilized seed by examining their outward appearance. Cycad growers use a test that involves dropping the seeds in water: viable seeds sink and lie on their longitudinal axes, while non-viable seeds (unfertilized and/or parasitized) float (Dehgan & Yuen, 1983; Grobbelaar, 2004). The exceptions are E. arenarius, E. nubimontanus and E. transvenosus, which sink regardless of their viability.

Cycad seeds are typically shed when their embryos are still immature (i.e. proembryos), and it usually takes approximately 4-6 months of after-ripening for the proembryo to mature in order for the seed to germinate (Grobbelaar, 2004; Woodenberg et al., 2007). A study conducted on the post-shedding seed behaviour of E. villosus, E. gratus and E. natalensis by Woodenberg et al. (2007) showed that the seeds were shed at relatively high water contents (2.3-6.0 g [g.sup.-1]), and a drying trial confirmed that they were recalcitrant. As a result of their recalcitrance, cycad seeds cannot be drieddown for long-term storage. Many South African Encephalartos species shed their seeds in early summer, but due to their need for after-ripening, are unable to germinate before the end of the summer rainy season, and must therefore persist for several more months through the dry winter until favourable conditions for germination return the following spring/summer. In most cases the viability of Encephalartos seeds starts to decrease after 2 years of storage in open containers at room temperature, although some can remain viable for up to 5 years (Grobbelaar, 2004). While the longevity of Encephalartos seeds in the wild has not been documented, it is likely to be shorter than that of seeds stored under less-harsh ambient conditions indoors.

For successful germination of cycad seeds under nursery conditions it is recommended that the dormant seeds are stored in open containers or coarsely-woven plastic bags and that the seeds be immersed in water for 48 h monthly to prevent desiccation (Grobbelaar, 2004). Before seeds are stored, their fleshy sarcotestae are removed by keeping them moist in plastic bags for a few days and then nibbing the sarcotestae off (Grobbelaar. 2004). In cultivation, it is advisable to keep the seeds moist throughout the gennination period, but in the wild, it has been noted that the seeds of some species (especially those adapted to fairly arid environments, e.g., E. lehmannii, E. horridus and E. cycadifolius) are capable of genninating on bare ground or exposed rock surfaces in full sun (Dyer, 1965). These germinating seeds are however highly unlikely to establish. Donaldson (1995b) noted that Encephalartos seeds in the wild generally exhibit high mortality rates, and field observations have revealed that in the rodent-dispersed grassland cycad E. cycadifolius, virtually all seeds that are not taken underground into rodent caches die. Experimental reintroduction of E. tniddelburgensis seedlings to the wild showed that seedlings planted in protected sites among rocks had a higher survival rate than those planted out in the open (Rousseau & Rousseau, 2011). For successful gennination and establishment, it appears that the seeds of most, if not all, Encephalartos species require sheltered microsites between rocks or under the shade of other plants where they are protected from desiccation and heat stress due to harsh direct sunlight and fire.

Ethnobotany

Traditional use of cycads by humans dates back many centuries, and throughout the world, cultural groups living in regions where cycads occur regard them as an important food source, especially during times of famine (Thieret 1958; Beaton 1982; Whitelock, 2002; Ravele & Makhado, 2008). The two most commonly consumed cycad plant parts are the seeds and the pith from the stems, which are both rich in starch, but also contain toxins which need to be eliminated prior to consumption (Thieret 1958; Beaton 1982; Bonta et al., 2006; Radha & Singh 2008; Ravele & Makhado, 2008). Other uses of cycads have included medico-magical practices, production of gum and fibre, and decoration during special occasions (Thieret, 1958). Since the mid- 1990s there has been a growing interest in cycad ethnobotany, with intensive studies on species in Central America (e.g., Bonta et al., 2006; PerezFarrera & Vovides, 2006; Bonta, 2010), India (e.g., Varghese & Ticktin, 2006; Radha & Singh, 2008; Krishnamurthy et al., 2013) and South Africa (Osbome et al., 1994; Crouch et al., 2003; Ravele & Makhado, 2008, 2009; Cousins et al., 2011, 2012, 2013; Williams et al., 2014; Williamson et al., 2016; Bamigboye et al., 2017). The traditional use of cycads in African countries other than South Africa is, however, poorly known, particularly which species are used, the quantity harvested (and traded), and the actual uses of the taxa concerned.

The use of African Cycads for Traditional Medicine

Encephalartos Species

In South Africa, Encephalartos species and Stangeria eriopus (respectively referred to as isiGqiki-somkovu or isiDwaba-somkovu, and imFingo in Zulu) form part of an extensive trade in indigenous plants for traditional medicine (muthi), which was valued at USD 412 million/year in 2007 (Mander et al., 2007). Encephalartos species are sold in South African muthi markets primarily in the form of 'bark' strips (leaf bases attached to an underlying layer of cortex tissue) as well as sections of whole stems (Cousins et al., 2011). Since Encephalartos plant parts do not appear to have many chemical compounds with therapeutic properties (Hutchings et al., 1996; Arnold et al., 2002), Encephalartos species are chiefly used for magical purposes, which typically involves averting the malicious activities of evil spirits (Crouch et al., 2003). In terms of treating physical ailments with Encephalartos species, Ravele & Makhado (2009) report that the roots, bark and leaves of E. transvenosus are boiled in water and consumed as a cure for stomach pain, heart attacks and strokes. While the leaves of E. transvenosus are known to contain macrozamin, a carcinogenic glycoside of methylazoxymethanol (Nair & Van Staden, 2012), it is unclear whether this compound is the active ingredient in the remedies prepared using E. transvenosus--or any other Encephalartos species--for medicinal use.

Since Encephalartos species traded for muthi in South Africa are sold primarily in the form of stem material, positively identifying the species in the absence of diagnostic plant parts (leaves and cones) presents a major challenge (Cousins, 2012; Cousins et al., 2012, 2013). The last decade has seen substantial progress in the identification of cycads (including Encephalartos species) using DNA barcoding techniques (Sass et al., 2007; Prakash & Van Staden, 2008; Rousseau, 2011), but until a recent study by Williamson et al. (2016), these methods were not feasible for rapidly identifying stem samples in traditional medicine markets. Another layer of complexity is added when attempting to quantify the cycad stem material traded in the markets and hence extracted from the wild (Cousins et al., 2011). Since Encephalartos stems are sold as fragments, and almost never whole, it is impossible to determine the exact number of plants impacted by harvesting (Cousins et al., 2011), although reasonable estimates can be made (see Williams et al., 2014).

Cousins et al. (2011) quantified the trade in South African Encephalartos species in Johannesburg and Durban--the cities with the country's two largest muthi markets. The quantity of Encephalartos stem material sold in the markets was estimated, as were the diameters of the stems from which the fragments purchased from the traders were harvested (Cousins et al., 2011). An estimate of 9.0 metric tons of Encephalartos traded in the Durban market in 2009 (Cousins et al., 2011) indicated that the trade is more extensive than suggested by earlier market-based ethnobotanical studies (Cunningham, 1988; Williams et al., 2001; Williams, 2003). Photographs taken over three time-steps (1940s, mid-1990s and 2014), at populations of 20 South African Encephalartos species revealed that harvesting for traditional medicine was the second largest cause of population decline (illegal removal for the horticultural trade being the largest) (Okubamichael et al., 2016). Cousins et al. (2011, 2012) showed that most harvesting appears to be from sub-adult and adult cycads, and that bark strips are harvested from large arborescent plants (Fig. 3d), whereas the entire plant is removed in the case of smaller individuals and subterranean-stemmed species (e.g., E. villosus and E. ngoyanus). Bark harvesting on a small scale does not appear to impact significantly on the plants, but ongoing intensive bark stripping can result in substantial declines, and in some cases, complete decimation of subpopulations (e.g. E. friderici-guilielmi at Tsolo in the Eastern Cape, South Africa) (Okubamichael et al., 2016). In the case of subterranean species, the removal of whole plants effectively eliminates them from the population, and hence impacts on population persistence are likely to be much more severe.

In the absence of diagnostic plant parts, Cousins et al. (2012) took an approach to species identification which involved the use of stem moiphological characters and harvesting localities cited by the traders from whom the stem samples were purchased. The geographical distributions of the Encephalartos species occurring in South Africa were compared with the harvesting localities, and possible species were short-listed based on overlaps between the distributions and the harvesting areas. The stem material was then examined with reference to the species descriptions of the short-listed species in the literature, and samples compared with one another, following which positive identifications could be made (Cousins et al, 2012). A literature review in the same study indicated that 25 South African Encephalartos species are used for traditional medicine, including six Critically Endangered species (but not all are necessarily traded). The results from the market surveys showed that the species most commonly traded in Johannesburg and Durban's two largest traditional medicine markets (in order of abundance) were E. natalensis, E. villosus and E. ghellinckii, with smaller quantities of E. ngoyanus and what appeared to be E. senticosus and E. ferox (Cousins et al., 2012). A photographic identification guide to these six species presented in Cousins et al. (2013) may be used by ethnobotanists and conservation practitioners to assist with rapid identification of the species used for traditional medicine.

Building on the results of Cousins et al. (2011, 2012), Williams et al. (2014) developed a technique to estimate the number of individual cycad stems from which the stem material had been harvested for traditional medicine. The method took into account the post-harvest age, physical condition, and stem diameter of the fragments that were identified to species level. An estimate of 81 different damaged stems (66% of which were likely to have been from E. natalensis) was calculated based on samples of 133 stem pieces from 56 traders. This calculation provided a significant advance in quantifying a previously unknown aspect of the trade in cycads for traditional medicine. Notwithstanding, marked surveys combined with quantitative assessments of wild populations known to be targeted by plant gatherers are required to provide more accurate measures of the extent and impact of the trade, and hence inform management decisions and enable the implementation of effective conservation strategies.

Williamson et al. (2016) conducted a follow-on study at the same two markets surveyed by Cousins et al. (2011, 2012), and used DNA barcoding to determine the diversity of the cycad species traded. An Encephalartos DNA barcode library was prepared and used to identify the stem samples purchased from the markets to species level. This library was assembled from 260 taxa (excluding market samples) representing 48 southern African species of known provenance that were identified by experts on the genus Encephalartos (Williamson et al, 2016). Using three different approaches to assign query market samples to known species (tree-based, similarly-based and character-based methods), a total of five Encephalartos species were identified (in order of prevalence): E. villosus, E. senticosus, E. natalensis, E. lebomboensis and E. ferox (Williamson et al., 2016). This study therefore confirms the presence in the markets of four of the species detected in previous studies, and indicates that an additional species (E. lebomboensis) is also being traded. The substantial impact of the traditional medicine trade on E. natalensis was also highlighted by Okubamichael et al. (2016). Populations of this species as well as those of E. friderici-guilielmi were disproportionately impacted by harvesting for traditional medicine compared to the other 18 species studied (Okubamichael et al., 2016). This study emphasised that an improved regulatory framework achieved through consultation and collaboration with traditional medicine practitioners is urgently needed to reduce cycad harvesting for traditional medicine to more sustainable levels.

Stangeria Eriopus

The underground stem of S. eriopus (often referred to as a caudex) has long been used traditionally by many ethnic groups in South Africa, especially the Xhosa and Zulu people (Watt & Breyer-Brandwijk, 1962; Osborne et al., 1994). S. eriopus caudices are used primarily as an ingredient in magical potions to protect against malevolent spirits, but also as an emetic. The toxins cycasin and macrozamin are both present in fresh and dried Stangeria stem material (Moretti et al, 1983); however, the emetic use thereof is ascribed to its high concentration of sodium sulphate (Osborne et al., 1994). The caudices are usually sliced before being sold, but whole plants, or multi-species mixes (intelezi) which include S. eriopus are also traded. In a study of the trade in S. eriopus in two traditional medicine markets in South Africa's KwaZulu-Natal province, Osborne et al. (1994) showed that a total of 3410 plants (2380 kg of stem material) were sold in a single month in 1992 in the two markets combined. This quantity excludes that traded at other markets and retail outlets in South Africa, which would probably have increased it considerably (Osbome et al., 1994). Furthermore, in a case study of the muthi trade in KwaZulu-Natal, Mander (1998) ranked the popularity of S. eriopus among the top 15% of the ca. 400 species traded.

Stangeria eriopus is also traded in the Faraday muthi market and retail outlets on the Witwatersrand, Gauteng Province, South Africa (Williams et al, 2007). Fifty-six percent of the 50 muthi shops surveyed in 1995 sold S. eriopus, with an estimated 2.1 [+ or -] 1.3 (mean [+ or -] S.D.) 50 kg-size maize bags stocked per shop per annum. Nine percent of a sample of 100 traders in the Faraday market in Johannesburg sold 5. eriopus, with an estimated thirty-eight 50 kg-size maize bags purchased annually by all the traders in the market combined (n = 164) (Williams et al., 2007). The muthi trade continues to pose a significant threat to wild populations of S. eriopus, many of which have also been lost due to agricultural activities, especially the establishment of pineapple farms in the Eastern Cape and the sugar cane industry in KwaZulu-Natal (Vorster & Vorster, 1985). In an endeavour to save S. eriopus from extinction in the wild, the Stangeria eriopus Conservation Project was launched at the Durban Botanic Gardens in 2000 (Crouch et al., 2000; Douwes, 2002; Douwes et al., 2004). (See subsequent section on ex situ conservation of African cycads for further details.)

Other Ethnobotanical Uses of African Cycads

The pith of Encephalartos stems is known to have been used to make flour for bread during times of famine in South Africa (several species), Zimbabwe (E. manikensis), Mozambique (E. ferox and E. manikensis) and Zanzibar (E. hildebrandtii) (Thieret, 1958; Dyer, 1965; Donaldson, 2003). This practice was first observed by Thunberg in the early 1770s when travelling through parts of southern Africa (Thieret, 1958). The pith was typically scooped out of the stems and buried underground for approximately two months and allowed to ferment. It was then exhumed, kneaded with water and shaped into small loaves or cakes to be baked on open fires (Thieret, 1958). Observations of these practises by Dutch colonists in South Africa gave rise to the name "broodboom" (literally, "bread tree") (Thieret, 1958), which is still applied to cycads by the Afrikaans-speaking community in South Africa today. It is largely unknown to what extent African cycads are still used in this manner, but Donaldson (2003) suggests it is probably only restricted to times of famine, since the plants generally occur in low numbers, they grow very slowly, and are often too inaccessible to make them suitable for frequent use.

In addition to Encephalartos stems, their leaves and seeds are also reported to have been used as a food source. Stiles (1981) reported that in Kenya, the Boni people regarded the seeds of E. hildebrandtii as their most important wild plant food. The cones were harvested in August/September and the seeds were then dried, pounded, and made into bread or gruel (Stiles, 1981). It appears that the Wasanya people of Kenya have been observed using E. hildebrandtii in a similar manner in more recent times as well (Whitelock, 2002). On the islands of Zanzibar and Pemba, starch from E. hildebrandtii seeds and stems has been used to make a porridge called ugali (Whitelock, 2002). Also in Zanzibar, people used the young foliage of E. hildebrandtii and/or C. thouarsii during a food crisis in the early 1970s (Nowak & Lee, 2011). Giddy (1984) reported that cycad seeds in the Eastern Cape province of South Africa have been used by local tribes as a seasonal dietary supplement. A similar practise is still observed in South Africa's Limpopo Province by people living in close proximity to populations of E. transvenosus (Ravele & Makhado, 2009). This species is also prized as an ornamental plant in local households and public facilities, and both E. transvenosus and E. villosus planted in homesteads are believed to repel evil spirits and lightning (Ravele & Makhado, 2009; Zobolo & Mkabela, 2006). The leaves of E. transvenosus are also used by local people as roofing material (Bamigboye et al., 2017), and to decorate tables at special occasions (Ravele & Makhado, 2009). Bamigboye et al. (2017) documented what appears to be a new usage of E. transvenosus, which involves grinding the inner part of the bark into a powder and then mixing it with other ingredients to be sniffed or smoked as hard drugs. Upon a visit to a population of E. tegulaneus in Kenya, Heenan (1977) noted that a number of large decumbent live specimens of this species had been partially hollowed out lengthwise by the local Sambum herdsmen to form cattle drinking troughs. It is not known if this practise continues today, and whether other Encephalartos species are used in a similar fashion.

Conservation Ecology

Approximately two-thirds of the world's cycad taxa are in danger of extinction, making cycads the most threatened group of plants globally (Hoffman et al., 2010; IUCN, 2016). The greatest threat posed to Africa's cycads is the illegal acquisition of wild plants by unscrupulous collectors for the horticultural trade (Donaldson & Bosenberg, 1999; Golding & Hurter, 2003; Donaldson, 2003, 2008; Okubamichael et al., 2016). For several decades, conservationists and botanists have repeatedly highlighted the plight of Africa's cycads, warning that the unabated disappearance of wild plants will ultimately result in species extinctions (Dyer, 1965; Osborne, 1990a; Grobbelaar, 1992; Giddy, 1995; Donaldson & Bosenberg, 1999; Golding & Hurter, 2003; Donaldson 2003, 2008; Coetzer, 2005). The problem has been of particular concern in South Africa, where it has assumed catastrophic proportions in recent years, and has consequently been termed the "South African cycad extinction crisis".

Conservation Status

The threatened status of African cycads has been revised in numerous assessments since the mid-1990s: (1) a global revision by Osborne (1995), (2) an assessment of the cycads in South Africa and Swaziland (Hilton-Taylor, 1996a,b), (3) as part of the 1997 IUCN Red List of Threatened Plants (Walter and Gillett, 1998), (4) an assessment of the rare and threatened plants of South Africa's KwaZulu-Natal province and neighbouring regions (Scott-Shaw, 1999), (5) a Red List account of African cycads as a whole (Golding & Hurter, 2003), (6) the IUCN Red Lists of 2003 (Donaldson, 2003) and 2009 (Raimondo et al., 2009), and (7) a review of the extinction risk of African cycads (Bamigboye et al., 2016). Of Africa's 67 cycad taxa, four are currently classified as Extinct in the Wild (6%) (E. brevifoliolatus, E. nubimontanus, E. relictus and E. woodii), 17 are Critically Endangered (CR) (26%), 10 are Endangered (EN) (15%), 16 are Vulnerable (VU) (24%), 13 are Near Threatened (19%) and only seven are Least Concern (LC) (10%) (IUCN, 2016) (Fig. 6). Hence, overall, four species no longer occur in the wild, 90% of the taxa are of Conservation Concern (NT, VU, EN, CR and EW) and two-thirds are threatened with extinction (VU, EN and CR). These data indicate that since the 2003 Red List assessment, two more species have become EW, the number of CR taxa has remained constant and the number of species listed as EN has risen from 10 to 12.

Major Threats to Wild Cycad Populations in Africa

Illegal Collecting

Identifying the causes of decline in African cycads is difficult, as there are few monitoring data for most cycad populations over a sufficiently long period (Donaldson, 2003). Nevertheless, a combination of data from cycad trade reports, permit records and two studies of matched photographs (Donaldson & Bosenberg, 1999; Okubamichael et al., 2016) provide an indication of the major causes (Donaldson, 2003). The most frequently cited driver behind the decline of African cycad populations is the trade in wild-collected plants (Dyer, 1965; Giddy, 1984; Goode, 1989, 2001). In South Africa, the illegal removal of cycads from the wild for the horticultural trade is the primary cause of the drastic declines witnessed in many species, and these declines have accelerated substantially over the last two decades (Donaldson & Bosenberg, 1999; Donaldson, 2008; Okubamichael et al., 2016). In a study of matched photographs of 20 South African Encephalartos species (a total of 626 cycad stems) over three time steps (1940s, mid-1990s and 2014), Okubamichael et al. (2016) showed that while 78% of the cycads photographed in the 1940s remained in the mid-1990s, by 2014 this proportion had fallen to only 16%. A somewhat unexpected finding was that cycad removal was greatest on both privately-owned and conserved land compared to communal land (Okubamichael et al., 2016). The highest losses of cycads on communal land resulted from harvesting of stem material for traditional medicine (Okubamichael et al.. 2016).

It should also be borne in mind that the threatened status of Africa's cycads is partly due to the natural rarity and decline of several species, e.g., E. dolomiticus, E. dyerianus, E. latifrons and E. powysii (Donaldson, 2003). Encephalartos latifrons was reported to be already scarce at a time when large-scale collecting for botanical gardens was only starting to take place, and several decades prior to the intensive decimation of cycad populations by private collectors (Pearson, 1916). While natural rarity is certainly a factor, it is clear that illicit acquisition of wild cycads by collectors has greatly accelerated the demise of many species (Donaldson. 2003, 2008). A classic example is E. relictus (EW)--known from only one small colony in Swaziland--so named because by the time of its description in 2001, it was already classified as Extinct in the Wild (Grobbelaar. 2004).

Cycads appeal to collectors due to their striking and often impressive appearance, their status as "living fossils" of the plant world, and their high monetary value owing to their rarity, aesthetic qualities and slow growth rates (Osborne, 1990a; Goode, 2001; Grobbelaar, 2004). Certain cycad collectors exhibit a compulsion to possess a representative of every species, and will go to great lengths to acquire the taxa they desire, sometimes even using sophisticated machinery (e.g., Vice, 1995) including helicopters to remove large individuals plants from the wild (Grobbelaar, 2004). This proclivity is taken a step further when individual species comprise different "variants" that come from separate localities and show differences in morphological features, especially the foliage (e.g., the 11 variants of E. nubimontanus sensu De Klerk, 2004). Targeted collecting of such species therefore propels them to extinction not only at certain populations in parts of their range, but at many, if not all the variant populations across their distribution. Cycads with a combination of natural rarity and striking and unusual physical appearances have tended to be the most severely affected (e.g., E. cupidus, E. heenanii, E. hirsutus, E. inopinus and E. nubimontanus). Cycad theft in Africa is not restricted to wild populations only: on numerous occasions cycads in several botanical gardens in South Africa have also been stolen by private collectors (K. Van der Walt, pers. comm.).

Another compounding factor in the extirpation of wild cycad populations in Africa is the socio-economic status of the local inhabitants of the regions where the cycads occur. Giddy (1995) highlights how an E. natalensis population in KwaZulu-Natal, South Africa was a source of income in an economically depressed rural area where unemployment and a lack of skills and education were a pervasive problem. The cycads were removed by the local people and sold to collectors, and were hence viewed as a natural resource that could be liquidated to pay for basic necessities such as housing and firewood (Giddy, 1995). Unscrupulous collectors are known to exploit the impoverished state of the local inhabitants by appointing them to remove cycads from the wild for a meagre wage after which the plants are sold for astronomical prices by comparison.

Habitat Loss

African cycad populations are often localized, and dispersal usually occurs over relatively short distances. Consequently, habitat destruction could potentially be one of the greatest threats to cycad populations, as entire populations, or even species, could be extirpated in one land-clearing event (Donaldson, 2003). Indeed, there are numerous reports of declines in wild populations resulting directly from habitat destruction or transformation. The South African grassland cycads E. laevifolius (CR) and E. humilis (VU) are threatened by afforestation with exotic pines and eucalypts, often surviving only in habitat islands between the plantations (Goode, 2001; Donaldson, 2003). Coastal resort developments and urban expansion have destroyed populations of E. hildebrandtii (NT) in Zanzibar, and populations of E. horridus (EN) and E. altensteinii (VU) in South Africa (Goode, 2001; Donaldson, 2003). Bush clearing for agriculture has altered or directly depleted populations of the Kenyan cycad E. kisambo (EN), E. gratus (VU) in Mozambique, E. schmitzii (VU) in Zambia, as well as E. arenarius (EN) and E. latifrons (CR) in South Africa (Donaldson, 2003; Daly et al., 2006). It appears that in these cases the people involved in land clearing for agriculture and developments are unaware of, or indifferent to, the threatened status of the cycads, as well as their monetary value, and therefore make little or no effort to preserve the plants in situ or conduct relocations. Notwithstanding, a concerted effort was made to remove ca. 6000 E. senticosus (VU) plants from the construction site of the Jozini dam on the border of South Africa and Swaziland in order to conserve the plants ex situ (Grobbelaar, 2004). In another rescue operation, 6000 E. humilis plants were relocated from an area in South Africa's Mpumalanga Province that was destined for afforestation (Boyd, 1995).

An important consideration is that habitat destruction may also indirectly affect cycad populations by making them more visible when surrounding vegetation is removed, and hence more accessible to collectors (Donaldson, 2003). Furthermore, fragmentation of cycad populations due to habitat destruction and/or collecting may also be indirectly deleterious due to reductions in pollinator numbers and increased distances between plants, making pollination less likely to occur. This phenomenon is particularly evident in the case of E. latifrons and E. middelbergensis populations, where natural pollinators appear to be locally extinct, thus resulting in zero seedling recruitment (Goode, 2001; Daly et al., 2006). Inbreeding depression and diminished resilience to stochastic events such as droughts and floods are also potential problems resulting from fragmentation. A study of matched photographs by Donaldson & Bosenberg (1999) showed that declines in Encephalartos populations due to habitat destruction in southern Africa occurred at ca. 12% of the sites surveyed, which was greatly overshadowed by the extent of declines caused by illicit collecting (60%). However, in African countries further north habitat destruction appears to be the biggest threat to the persistence of wild cycad populations (Golding & Hurter, 2001). It is, however, difficult to measure the overall impact of habitat destruction (Donaldson, 2003), and intensive fieldwork in countries north of southern Africa is required to assess the threats to Encephalartos populations in order to implement suitable conservation strategies.

Invasive Species

In southern Africa, invasive alien vegetation appears to have caused population declines in relatively few instances (2% of Encephalartos sites surveyed by Donaldson & Bosenberg (1999)). Nonetheless, invasive alien plants are present in numerous regions where cycads occur and in some instances they pose a potential threat to extant cycad populations (Donaldson, 2003). Dense stands of Lantana camara L., for example. have invaded farmlands in the Eastern Cape Province of South Africa where E. princeps (VU) and S. eriopus (VU) occur. Apart from weeds smothering seedlings and juveniles, chemical control of the alien species using herbicides may also destroy the cycads if selective spraying is not conducted (Donaldson, 2003). Another example is the stands of guava (Psidium guajava L.) that have become so dense in parts of Swaziland that it was impossible to locate cycads photographed during the matched photography study by Donaldson & Bosenberg (1999). Other alien plant species that are invasive in areas where cycads occur include prickly pear, (Opuntia ficus-indica L. (Mill.)), and Acacia (wattle) species (Donaldson, 2003).

The foremost impact of alien plants on cycad populations will likely be on recruitment rates owing to reduced coning frequencies due to shading and the transformed environment for germination and recruitment (J.S. Donaldson, unpublished data). Observations of a population of E. humilis in the Mpumalanga Province, South Africa, revealed that adult cycads persisted in the deep shade of large invasive alien Acacia mearnsii trees, but were totally leafless in midsummer when other more exposed E. humilis individuals were in leaf, and would therefore most likely also exhibit reduced coning frequencies (S.R. Cousins, pers. obs.). Quantitative data on the impacts of alien invasions on cycad populations may not reflect any negative impacts initially, but the long-term effects would probably be negative and will need to be carefully monitored (Donaldson, 2003).

Conservation Strategies

The alarming declines observed in a large number of African cycad populations have not always resulted from a lack of conservation planning and interventions. Detailed conservation strategies have been formulated for African cycads at several levels: the global cycad status survey and conservation action plan (Donaldson, 2003), and conservation plans for (1) South African cycads (Osborne, 1990a; Grobbelaar, 1992; Donaldson, 2008), (2) species in South Africa's KwaZulu-Natal Province (Anonymous, 2004), (3) single species Red Listed as Critically Endangered (E. latifrons in Daly et al. (2006)), and (4) imperilled populations of less-threatened species (E. natalensis (NT) in Giddy (1995)). South Africa has spearheaded the conservation of cycads in Africa, by establishing reserves, developing legislation, assigning resources to law enforcement and monitoring, using novel technologies such as microchips, microdots and the installation of elaborate security systems in botanical gardens where large cycad collections are housed (Daly et al., 2006). Cycads have undoubtedly benefited from many of these conservation measures, and if success were to be measured in terms of the numbers of reserves containing cycads, the number of regulations that restrict trade in cycads, or the effort spent on law enforcement, then cycads would be high on the list of plant conservation success stories (Donaldson, 2004b). Yet cycad populations continue to decline, and a more holistic approach to conservation, which goes beyond a simplified approach of establishing reserves and implementing restrictive legislation, is crucial (Donaldson, 2004b).

Trade Restrictions

CITES (the Convention on International Trade in Endangered Species of Wild Fauna and Flora) is the best known and most widely applied mechanism for regulating the trade in endangered species. All Encephalartos species and Stangeria eriopus are included in CITES Appendix I, which means that the international trade in wild specimens (or parts thereof) for commercial purposes is prohibited, but trade in artificially-propagated plants is permitted (Donaldson, 2003, 2004c). CITES permits are obligatory for all imports and exports of cycads and cycad material. In South Africa, where the illicit trade in cycads has been particularly rife, strict legislation has been implemented which requires permits for the possession and translocation of all indigenous cycads or cycad material. Since May 2012, it is prohibited to harvest, trade, sell, buy, donate, import, export, convey or receive any wild-collected cycads in South Africa, and the possession of a cycad of wild origin is prohibited, unless it forms part of legally-obtained parental stock for which a permit was issued prior to May 2012 (Retief et al., 2015). Harvesting of wild cycads in South Africa without a permit has been illegal since the 1970s, but unfortunately, the strict legislation intended to protect wild cycads (in South Africa and elsewhere in Africa) has been largely ineffective, as law enforcement agencies have lacked capacity to curb illegal removal of wild plants (Donaldson, 2003; Retief et al., 2014). Even when cycad poachers are apprehended, they may avoid prosecution because of technicalities in the law that require proof from the state that the plants were indeed wild-collected (Donaldson, 2003, 2008). However, the new advances being made in the use of forensic tools to determine the origin of cycad material shows promise in dealing with this issue (Nordling, 2014; Retief et al., 2014, 2015).

In Situ Conservation

One of the greatest challenges facing cycad conservation in situ in Africa is that many species are simply free for the taking, as emphasized by Goode (2001). Only approximately 25 African cycad taxa (37%) are included in one or more reserves, and 13 Critically Endangered, four Endangered and eight Vulnerable species do not appear to occur in any protected areas (Donaldson, 2003). Unfortunately, even when cycads are protected within a reserve, they may still be plundered by cycad collectors (as was the case with E. dyerianus (CR) at the Lillie Cycad Reserve in Limpopo Province, South Africa, where >100 plants were illegally removed in January 2008 (Donaldson, 2008)). The altitudinal position and remoteness of cycad populations often plays an important role in determining the extent to which they are impacted by illicit harvesting. The two forms of E. ghellinckii, for instance, occur separately in low- and high-lying parts of the Drakensberg Mountains in KwaZulu-Natal, South Africa (Goode, 2001). Populations of the so-called montane form are largely inaccessible due to steep slopes, making the uprooting and transportation of large, heavy plants difficult (Goode, 2001). By contrast, colonies of the lowland form are more accessible, and tend to be more heavily plundered by both collectors and medicinal plant gatherers (Goode, 2001). A number of African cycads occur in very inaccessible, mountainous terrain (e.g., E. cycadifolius, E. sclavoi and E. turneri), thus deterring collectors; notwithstanding, species occurring on mountains may still be heavily plundered, especially very popular taxa with exorbitant monetary value, e.g., E. hirsutus (Goode, 2001; Grobbelaar, 2004). By contrast, the South African grassland cycad, E. lanatus occurs in fairly accessible areas, but seldom survives transplantation (Dyer, 1965), and is thus less-frequently targeted by collectors.

In some cases, local communities residing close to cycad populations that they consider valuable protect the plants by prohibiting collectors from removing them (e.g., E. bubalinus, E. gratus and E. transvenosus) (Miringu, 1999; Goode, 2001). Some farm owners who have cycads on their properties are supportive of cycad conservation and make an effort to prevent illegal collecting on their farms (e.g., E. dolomiticus) (Goode, 2001). Moreover, a few strongly conservation-minded land owners have taken initiative in artificially pollinating plants in wild populations and propagating the resultant fertile seeds for reintroductions (e.g., E. middelburgensis and E. cerinus) (Robbertse et al., 1999; Goode, 2001; Rousseau & Rousseau, 2011). Donaldson (2004b) suggested that the sale of seed harvested from wild populations (or plants propagated from these seeds) by land owners may be permitted when it is combined with artificial pollination and actions that prevent illegal harvesting of plants. A condition would be that a proportion of the seeds be left in the wild or used to plant out seedlings to increase chances of population growth (Donaldson, 2004b). This arrangement would not only incentivise land owners to safeguard the cycads on their properties, but would also facilitate recruitment in populations exhibiting reproductive failure due to loss of pollinators and/or wide separation of male and female individuals because of gaps created by collectors.

In order to ensure the persistence of threatened cycads with critically small populations in situ, and re-establish wild populations of Extinct in the Wild taxa, reintroductions and augmentation using material from garden plants is often the only option (Daly, 2006; Da Silva et al., 2012). This kind of intervention can be achieved by introducing seed and/or seedlings propagated from cycads in botanical gardens or artificially-pollinated wild plants, or by relocating suckers from garden plants with multiple stems (Da Silva et al., 2012). Augmentation has proved to be successful in the past, when South Africa's former Transvaal nature conservation authority reinforced existing populations of E. cupidus and E. dyerianus with 300 artificially propagated seedlings each (Boyd, 1995). After 10 years of monitoring, a survival rate of 92% was achieved for E. cupidus. The survival rate for E. dyerianus was also high at 80%, although the number of years of monitoring was not specified (Boyd, 1995). Da Silva et al. (2012) caution, however, that if material from garden plants is to be used for reintroductions or augmentations, it must first be determined whether plants and/or propagules from different sources can be mixed without compromising the genetic structure of the population. Results from a study of the extent and structure of genetic diversity in wild and ex situ populations of the Critically Endangered South African cycad E. latifrons suggested that remnant subpopulations of the species were originally a single panmictic population, with historically high levels of gene flow (Da Silva et al, 2012). This implies that suckers, seedlings, and pollen can be transferred between populations without reducing fitness (Da Silva et al., 2012). The study also underlined the importance of ex situ collections, even those of unknown provenance. Da Silva et al. (2012) concluded that where collecting is a major cause of decline in a particular cycad species, important genetic material may be found in public and private collections where the exact provenance is not always known; however, if the plants are at least known to be of wild origin, then genetic techniques can be used to assess their potential contribution to restoration efforts.

Ex Situ Conservation

Fortunately, no African cycad species are known to have gone extinct subsequent to their discovery, as they are all conserved ex situ either in botanical gardens or private collections. Numerous cycad nurseries trading in strictly legal cycads have also been established by nature conservation authorities in South Africa in an effort to satisfy the horticultural demand for cycads. Indeed, there is usually less pressure on wild populations of species that are available in large numbers in cycad nurseries (e.g., E. senticosus) (Grobbelaar, 2004). Six South African botanical gardens have also established living collections of Encephalartos species for ex situ conservation, notably the Lowveld National Botanical Garden in Nelspruit, Kirstenbosch in Cape Town and the Durban Botanic Garden (Donaldson, 2003). African cycads are also fairly well represented in botanical gardens in Zimbabwe and Kenya (Donaldson, 2003). These living collections serve as seed orchards for propagation and reintroductions, and are also used for education and research. In addition to a display collection of cycads at the Lowveld National Botanical Garden is a collection of a large number of South African Encephalartos species of known provenance in a secure, fenced-off area (Walters, 2003). This seed orchard and living gene bank is specifically maintained to ensure that reintroductions (by seed or plants) can be conducted using appropriate (site-specific) genetic material (Walters, 2003). This pioneering work led to the establishment of similar collections in other parts of the world.

The Stangeria eriopus Conservation Project which was launched at the Durban Botanic Gardens in 2000 has been exemplary in terms of ex situ cycad conservation (Crouch et al., 2000; Douwes, 2002; Douwes et al., 2004). The aim of the project was to collect samples of live plants from as many S. eriopus populations as possible to establish a genetically representative ex situ living collection (Douwes, 2002). The S. eriopus collection was made available for scientific study on the species' taxonomy, pollination and seed biology, as well as for micropropagation protocol development and re-introduction into the wild in protected areas (Douwes et al, 2004). By establishing a large living collection of S. eriopus plants ex situ, the plants are much more accessible for such studies, and artificial pollination also provides a steady supply of seed for further propagation, sale to the horticultural trade, and for studies on seed banking techniques (Douwes et al., 2004). Thousands of African cycads are also maintained in private collections and botanical gardens internationally, some of which are of known origin at the population level. It is important that conservation authorities collaborate with conservation-minded cycad collectors who possess rare and threatened Encephalartos species to make use of seeds from these plants for reintroductions and to augment living collections in botanical gardens.

Alleviating the South African Cycad Extinction Crisis

In response to the cycad extinction crisis in South Africa, the South African National Biodiversity Institute (SANBI) has spearheaded the development and implementation of a detailed conservation strategy for the country's cycads. It is a collaborative initiative between SANBI, the Conservation Breeding Specialist Group (CBSG), the Cycad Specialist Group of the IUCN/SCC, The Endangered Wildlife Trust (EWT), TRAFFIC, various universities, and national government. The initiative has involved working with cycad collectors and owners of cycad nurseries to promote legal artificial propagation of cycads for the horticultural trade and for re-introductions to wild populations. It has also included the South African Cycad Species Protection Project, which focused on providing skills and knowledge to 194 law enforcement officials to enhance their capability to curb the illicit cycad trade (Rayner & Pires, 2016). The project included the development of an interactive photo-based electronic species identification application--the SANBI Cycad Identification Tool--for use by enforcement officials and the general public (Rayner & Pires, 2016). Furthermore, in order to meet their urgent need for conservation, a Biodiversity Management Plan for each of South Africa's 12 Critically Endangered cycad species has also been developed by SANBI with contributions from all relevant stakeholders. The main objectives of the initiative are to develop effective means of managing the horticultural trade and conserving the cycad species in situ, which includes re-introduction programmes for Critically Endangered and Extinct in the Wild taxa. Preliminary observations of reintroductions of E. middelburgensis (CR) in the Mpumalanga Province by seed broadcasting indicate that this is an effective means of reintroduction/ augmentation of wild populations (M.F. Pfab, pers. comm.).

Part of the approach to alleviating the South African cycad extinction crisis has involved research into the potential of forensic techniques such as radiocarbon dating and stable isotopes to prove the wild-origin of illegally-harvested cycads (Nordling, 2014; Retief et al, 2014, 2015). Microchipping of cycads in the wild has in some cases led to the successful prosecution of poachers and collectors, but this technique has significant disadvantages in that each cycad needs to be fitted with a microchip before the poaching event, and poachers are able to remove the microchips using X-ray scanners (Donaldson, 2008; Retief et al., 2014). Another technique that has been employed is DNA barcoding, which has been useful for confirming species identity and ascertaining which populations the illegally-obtained plants originated from, but failed to provide conclusive evidence of a wild versus cultivated origin of sampled collections (Retief et al., 2014). Two case studies by Retief et al. (2014), one each on E. lebomboensis and E. arenarius, indicate that changes in geographic locality (i.e. moving cycads from a locality in the wild to be planted in a garden hundreds of kilometres away) can be detected within the isotopic chronology of the cycads, providing a potentially powerful tool to ascertain the origin of illegally-collected cycads. The advantages of this technique are that it does not necessitate pre-marking of wild individuals, it cannot be removed or altered by the poacher/collector, and it can distinguish between wild versus cultivated origins for many years after the poaching event (Retief et al., 2014). It is hoped that when developed to its full potential, this method will offer an effective deterrent to members of the public from engaging in the illegal cycad trade (Retief et al., 2014).

Recommendations for Future Research

Studies on the population ecology of African cycads (particularly rare and threatened species) that include size/stage class distributions and estimates of population size and density (e.g., Lopez-Gallego, 2008) would provide important baseline data for the implementation of conservation and management interventions. Long-term population monitoring and the use of matrix population models (e.g., Raimondo & Donaldson, 2003; Perez-Farrera et al., 2006) are important for elucidating trends and making predictions about the future population ecology of cycad species, especially those undergoing plant extraction (e.g., Alvarez-Yepiz et al., 2011), reintroductions and/or population augmentation. More research is required on Encephalartos and Stangeria seed ecology, particularly dispersal and seed longevity in the wild (e.g., Burbidge & Whelan, 1982; Snow & Walter, 2007). Studies on the way in which fire affects African cycad populations--particularly grassland species--with respect to reproduction and population persistence (e.g., Grove et al., 1980; Negron-Ortiz & Gorchov, 2000) would be useful for informing the management of populations of these taxa. Finally, research is needed on the extent to which invasive alien plant species are impacting on Africa's cycad species and the specific impacts the invasive species have on population persistence.

DOI: 10.1007/s1222-017-9183-4

Acknowledgements Many thanks go to Dc Wet Bosenberg for producing the species distribution map. Wynand van Eeden is thanked for providing comments on the manuscript. Thanks go to John Donaldson. Pict Vorster, Gerhard van Deventer, Michael Calonje, Xander de Kock and David Muller for supplying photographs. We thank the University of the Witwatersrand, Johannesburg for providing funding.

Literature Cited

Alvarez-Yepiz. J. C., M. Dovciak, & A. Burquez. 2011. Persistence of a rare ancient cycad: Effects of environment and demography. Biological Conservation 144: 122-130.

Anonymous. 2004. A management plan for cycads in KwaZulu-Natal. Threatened Plant Conservation Unit. Biodiversity Conservation Advice Division, Ezemvclo KZN Wildlife.

Arnold. T. H., C. A. Prentice, L. C. Hawker, E. E. Snyman, M. Tomalin. N. R. Crouch & C. Pottas-Bircher. 2002. Medicinal and Magical Plants of Southern Africa: an Annotated Checklist. Strelitzia 13. National Botanical Institute. Pretoria.

Bamigboye. S. O., P. M. Tshisikhawe & P. J. Taylor. 2016. Review of extinction risk in African cycads. Phyton International Journal of Experimental Botany 85: 333-336.

Bamigboye. S. O.. P. M. Tshisikhawc & P. J. Taylor. 2017. Detecting threats to Encephalartos transvenosus (Limpopo cycad) in Limpopo province. South Africa through indigenous knowledge. Indian Journal of Traditional Knowledge 16(2): 251-255.

Bayliss, J., C. Burrow. S. Martell & H. Staude. 2009. An ecological study of the relationship between two living fossils in Malawi: the Mulanje Tiger Moth (Callioratis grandis) and the Mulanje Cycad (Encephalartos gratus). African Journal of Ecology 48: 472-480.

Beaton, J. M. 1982. Fire and water: Aspects of Australian Aboriginal management of cycads. Archaeology in Oceania 17(1):51 58.

Bonta, M. 2010. Human geography and ethnobotany of cycads in Xi'ui, Tccnek, and Nahuatl communities of northeastern Mexico. Final Report. Cleveland.

Bonta, M., O. Flores Pinot, D. Graham, J. Haynes & G. Sandoval. 2006. Ethnobotany and conservation of Tiusinte (Dioon mejiae Standi. & L.O. Williams, Zamiaccae) in northeastern Honduras. Journal of Ethnobiology 26(2): 228-257.

Boyd, W. M. 1995. The translocation and re-establishment of priority Encephalartos species in the Transvaal. South Africa. Pp 423-434. In: P. Vorster (ed). Proceedings of the Third International Conference on Cycad Biology. The Cycad Society of South Africa, Stellenbosch.

Burbidge, A. H. & R. J. Whelan. 1982. Seed dispersal in a cycad, Macrozamia riedlei. Australian Journal of Ecology 7: 63-67.

Calonje. M., D. W. Stevenson & L. Stanberg. 2016. The World List of Cycads, online edition, 2013-2016. [cited 18 Feb 2016], Available from: http://www.cycadlist.org.

Chamberlain, C. J. 1919. The Living Cycads. The University of Chicago Press. Chicago.

Chamberlain, C. J. 1935. Gymnosperm Structure and Evolution. University of Chicago Press. Chicago.

Chang, D. C. N., N. Grobbelaar & J. Coetzee. 1988. SEM observations on cyanobacteria-infected cycad coralloid roots. South African Journal of Botany 54(5): 491 495.

Cilliers. A. 2012. Stimulation by fire of certain South African Encephalartos species. Encephalartos 110: 14-17.

Clarke, P. J.. M. J. Lawes, J. J. Midgley. B.-B. Lamont, F. Ojeda, G. E. Burrows, N. J. Enright & K. J. E. Knox. 2013. Resprouting as a key functional trait: how buds, protection and resources drive persistence after fire. New Phytologist. 197: 19-35.

Coctzer. I. A. 2005. The living fossils of Africa's biodiversity under threat: Can the most critically endangered Encephalartos cycads be saved from extinction? Enccphalartos 82: 14-17.

Cousins, S. R. 2012. The trade in South African Encephalartos species for traditional medicine: added pressure to the cycad extinction crisis. Encephalartos 107: 39-43.

Cousins, S. R., V. L. Williams & E. T. F. Witkowski. 2011. Quantifying the trade in South African Encephalartos spp. in the traditional medicine markets of Johannesburg and Durban, South Africa. Economic Botany 65(4): 356-370.

Cousins, S. R., V. L. Williams & E. T. F. Witkowski, 2012. Uncovering the cycad taxa (Encephalartos species) traded for traditional medicine in Johannesburg and Durban, South Africa. South African Journal of Botany 78: 129-138.

Cousins, S. R., V. L. Williams & E. T. F. Witkowski. 2013. Sifting through cycads: A guide to identifying the stem fragments of six south African medicinal Encephalartos species. South African Journal of Botany 84: 115-123.

Crouch, N. R., J. S. Donaldson, G. F. Smith, R. Symmonds, C. G. M. Dalzell & C. R. Scott-Shaw. 2000. Ex situ conservation of Stangeria eriopus (Stangeriaceac) at the Durban Botanic gardens, South Africa. Encephalartos 63: 16-24.

Crouch, N. R.. Smith, G. F., M. Lotter, & R. Symmonds. 2003. Encephalartos woodii--The first historically documented ethnomedicinal plant casualty in southern Africa. Encephalartos 73: 25-28.

Cunningham, A. B. 1988. An investigation of the herbal medicine trade in Natal/KwaZulu. Investigational report, Vol. 29. Pietermaritzburg: Institute of Natural Resources.

Da Silva, J. M., J. S. Donaldson, G. Reeves. & T. A. Hedderson. 2012. Population genetics and conservation of critically small cycad populations: A case study of the Albany cycad, Encephalartos latifrons (Lehmann). Biological Journal of the Linncan Society 105(2): 293-308.

Daly, B., J, S. Donaldson, Y. Friedmann, Q. Hahndiek, N. King, D. Newton, & A. Southwood. 2006. Albany cycad (Encephalartos latifrons) population and habitat viability assessment workshop report. Conservation breeding Specialist group (SCC/IUCN)/CBSG southern Africa. Endangered Wildlife Trust, Johannesburg.

De Klerk, D. 2004. Encephalartos nubimontanus: A distinction between 11 variants. Published by the author.

Dehgan. B. & C. K. K. H. Yuen. 1983. Seed morphology in relation to dispersal, evolution, and propagation of Cycas L. Botanical Gazette 144(3): 412-418.

Donaldson, J. S. 1993. Mast-seeding in the cycad genus Encephalartos: A test of the predator satiation hypothesis. Oecologia 94: 262-271.

Donaldson, J. S. 1995a. The Winterberg cycad: Surviving against the odds. Veld & Flora 81(2): 36-39.

Donaldson, J. S. 1995b. Understanding cycad life histories: An essential basis for successful conservation. Pp 8-13. In: J. S. Donaldson (ed). Cycad conservation in South Africa: Issues, priorities, and actions. Cycad Society of South Africa, Stellenbosch.

Donaldson, J. S. 1997. Is there floral parasite mutualism pollination in cycads? Pollination biology of Encephalartos villosus (Zamiaceac). American Journal of Botany 84(10): 1398-1406.

Donaldson, J. S. 1999. Insects associated with the cycads of Zimbabwe. Kenya and Zanzibar with comparisons to cycad insects from South Africa. Excelsa 19: 40-46.

Donaldson. J. S. 2003. Cycads: A status survey and conservation action plan. IUCN/SSC Cycad Specialist Group. Gland.

Donaldson, J. S. 2004a. Extinction of cycad pollinators--Do generalists or specialists survive as cycads decline? Pp 154. In: J. A. Lindstrom (ed). The biology, structure, and Systematics of the Cycadales. Proceedings of the Sixth International Conference on Cycad Biology. Nong Nooch Tropical Botanical Garden, Thailand.

Donaldson, J. S. 2004b. Seeds of hope: Can trade in cycad seeds help save cycads from extinction? The Cycad Newsletter: 11-14.

Donaldson, J. S. 2004c. CITES and cycad conservation: Away forward. The Cycad Newsletter: 8-11.

Donaldson, J. S. 2008. South African Encephalartos species. NDF workshop case studies: Case study 4: Encephalartos. Mexico.

Donaldson, J. S. & J. D. Bosenberg. 1995. Life history and host range of the leopard magpie moth, Zerenopsis leopardina Felder (Lepidoptera: Geometridae). African Entomology 3(2): 103-110.

Donaldson, J. S. & J. D. Bosenberg. 1999. Changes in the abundance of south African cycads during the twentieth century: Preliminary data from the study of matched photographs. Biology and conservation of cycad. In: C. J. Chen (Ed.) Proceedings of the Fourth international Conference on cycad biology. 240247. Beijing: International Academic Publishers.

Donaldson, J. S., I. Nanni & J. D. Bosenberg. 1995. The role of insects in the pollination of Encephalartos cycadifolius. Pp 423-434. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. The Cycad Society of South Africa, Stellenbosch.

Douwes, E. 2002. The Stangeria eriopus conservation project: Ensuring the survival of the Natal grass cycad. Veld & Flora 87(4): 162-163.

Douwes, E., N. R. Crouch, M. Mattson. C. G. M. Dalzell & G. F. Smith. 2004. The Stangeria eriopus conservation project, a gene-banking programme in action. Pp 85-91. In: J. A. Lindstrom (ed). The biology, structure, and Systematics of the Cycadales. Proceedings of the Sixth International Conference on Cycad Biology. Nong Nooch Tropical Botanical Garden, Thailand.

Dyer, R. A. 1965. The cycads of southern Africa. Bothalia 8: 405-515.

Fourie, S. P. 1995. Population census data and long-term monitoring of cycad populations. Pp 8-13. In: J. S. Donaldson (ed). Cycad conservation in South Africa: Issues, priorities, and actions. Cycad Society of South Africa. Stellenbosch.

Giddy, C. 1984. Cycads of South Africa. 2nd edition. C. Struik (Pty) Ltd. Publishers, Cape Town.

Giddy, C. 1995. Cycad conservation--a third world perspective. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. Cycad Society of South Africa. Stellenbosch.

Goode, D. G. 1989. Cycads of Africa. Struik Winchester.

Goode, D. G. 2001. Cycads of Africa Volume I. D & E Cycads of Africa Publishers. Sandton.

Golding, J. S. & P. J. H. Hurter. 2001. Perspectives on cycads in Africa. Encephalartos 68: 30-33.

Golding, J. S. & P. J. H. Hurter. 2003. A red list account of Africa's cycads and implications of considering life-history and threats. Biodiversity and Conservation 12: 507-528.

Grobbelaar, N. 1989. Disintegration of Encephalartos megastrobili. South African Journal of Botany 55(6): 581-585.

Grobbelaar, N. 1992. A more realistic conservation strategy for the south African cycads. Plant Life 6: 14-16.

Grobbelaar, N. 1999. Coming frequency, gender ratio, and pollination of Encephalartos transvenosus (Zamiaccae) at the Modjadji nature reserve, South Africa and the germination of this cycad's seed. Pp 309-318. In: C. Chen (ed). Biology and conservation of cycads. Proceedings of the Fourth International Conference on Cycad Biology. Beijing.

Grobbelaar, N. 2003. The effect of light strength on the survival of Encephalartos lanatus seedlings. Encephalartos 75: 21-23.

Grobbelaar, N. 2004. Cycads: With special reference to the southern African species. Published by the author. Pretoria.

Grobbelaar, N., W. Hattingh & J. Marshall. 1986. The occurrence of coralloid roots on the south African specics of the Cycadalcs and their ability to fix nitrogen symbiotically. South African Journal of Botany 52: 467-471.

Grobbelaar, N., J. J. M. Meyer & J. Burchmore. 1989. Coning and sex ratio of Encephalartos transvenosus at the Modjadji nature reserve. South African Journal of Botany 55(1): 79-82.

Grove, T. S., A. M. O'Connell & N. Malajczuk. 1980. Effects of fire on the growth, nutrient content and rate of nitrogen fixation of the cycad Macrozamia riedlei. Australian Journal of Botany 28: 271-281.

Harris, R. & W. Harris. 2003. Doomkop: Home of Encephalartos middelburgensis and E. lanatus. Encephalartos 73: 12.

Heenan, D. 1977. Some observations on the cycads of Central Africa. Botanical Journal of the Linnean Society 74: 279-288.

Heibloem, P. 1999. The Cycads of Central Africa. Palm and Cycad Societies of Australia, Brisbane.

Hilton-Taylor. C. 1996a. Red data list of southern African plants. Strelitzia 4. National Botanical Institute, Pretoria.

Hilton-Taylor, C. 1996b. Red data list of southern African plants: Corrections and additions. Bothalia 26: 177-182.

Hoffman, M., C. Hilton-Tayler, A. Angulo, M. Bohm, T. M. Brooks, S. H. M. Butchart, ..., S. N. Stuart. 2010. The impact of conservation on the status of the world's vertebrates. Science 330: 1503-1509.

Hurter, J. 1994. Focus on Encephalartos heenanii R.A. Dyer. Encephalartos 40: 4-7.

Hutchings, A., A. H. Scott, G. Lewis & A. B. Cunningham. 1996. Zulu Medicinal Plants: An Inventory. University of Natal Press. Pietermaritzburg.

IUCN Red List of Threatened Species. Version 2015-4. www.iucnredlist.org. Downloaded 8 January 2016. Jones, D. L. 1993. Cycads of the World: Ancient Plants in Today's Landscape. REED (William Heinemann). Chatswood. Australia.

Krishnamurthy, V., L. Mandle, T. Ticktin. R. Ganesan, C. S. Sancesh & A. Varghesc. 2013. Conservation status and effects of harvest on an endemic multi-purpose cycad, Cycas circinalis L., western Ghats. India. Tropical Ecology 54(3): 309-320.

Lopez-Gallego, C. 2008. Demographic variation in cycad populations inhabiting contrasting forest fragments. Biodiversity Conservation 17: 1213-1225.

Mander. M. 1998. Marketing of indigenous medicinal plants in South Africa: A case study in KwaZulu-Natal. FAO of the UN. Rome.

Mander, M.. L. Ntuli, N. Diederichs & K. Mavundla. 2007. Economics of the traditional medicine trade in South Africa. Pp 189-196. In: S. Harrison, R. Bhana, A. Ntuli. (eds). South African Health Review 2007. Health Systems Trust. Durban.

Marloth, R. 1914. Notes on the cntomophilous nature of Encephalartos. Transactions of the Royal Society of South Africa 4: 69-71.

Melville, R. 1957. Encephalartos in central Africa. Kew Bulletin 12: 237-257.

Miringu, B. W. 1999. The population and conservation status of the Kenyan cycads, and their potential as a horticultural crop. Pp 391-396. In: C. J. Chen. (ed). Proceedings of the Fourth international Conference on cycad biology. Beijing

Moretti, A., S. Sabato & G. G. Siniscalco. 1983. Taxonomic significance of methylazoxymcthanol glycosides in the cycads. Phytochcmistry 22: 115-117.

Nair, J. J. & J. Van Staden. 2012. Isolation and quantification of the toxic methylazoxymcthanol glycoside macrozamin in selected south African cycad species. South African Journal of Botany 82: 108-112.

Negron-Ortiz, V. & D. L. Gorchov. 2000. Effects of fire season and postfire herbivory on the cycad Zamia pumila (Zamiaccae) in slash pine savanna. Everglades National Park, Florida. International Journal of Plant Sciences 161(4): 659-669.

Nordling, L. 2014. Forensic chemistry could stop African plant thieves: Isotope analysis could help in the rush to save South Africa's cycads from extinction. Nature 514(17).

Norstog, K. J. & T. J. Nicholls. 1997. The Biology of the Cycads. Cornell University Press, Ithaca. N.Y.

Nowak, K. & P. C. Lee. 2011. Consumption of cycads Encephalartos hildebrandtii by Zanzibar red colobus Pivcolobus kirkii. Journal of East African Natural History 100: 123-131.

Oberprieler. R. G. 1995a. The weevils (Coleoptera: Curculionoidca) associated with cycads. I. Classification, relationships, and biology. Pp. 295-334. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. Cycad Society of South Africa, Stellenbosch.

Oberprieler, R. G. 1995b. The weevils (Coleoptera: Curculionoidca) associated with cycads. 2. Host specificity and implications for cycad taxonomy. Pp 335-365. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. Cycad Society of South Africa, Stellenbosch.

Oberprieler, R. G. 2004. "evil weevils"--The key to cycad survival and diversification? Pp 170-194. In: J. A. Lindstrom (ed). Proceedings of the sixth international cycad Conference on cycad biology. Nong Nooch Tropical Botanical Garden, Thailand.

Okubamichael, D. Y., S. L. Jack, J. D. Bosenberg, M. T. Hoffman & J. S. Donaldson. 2016. Repeat photography confirms alarming decline in south African cycads. Biodiversity Conservation 25(11): 2153-2170.

Osborne, R. 1990a. A conservation strategy for the south African cycads. South African Journal of Science 86: 220-223.

Osborne. R. 1990b. Two new reports of cycad sex changes. Encephalartos 23: 18-20.

Osborne, R. 1995. The world cycad census and a proposed revision of the threatened species status for cycad taxa. Biological Conservation 71: 1-12.

Osborne, R. & Gorelick, R. 2003. Sex change in cycads. Palms & Cycads 76: 10-15.

Osborne, R. & Gorelick, R. 2007. Sex change in cycads: Cases, causes, and chemistry. Memoirs of the New York Botanical Garden 97: 335-345.

Osborne, R., A. Grove, P. Oh, T. J. Mabry, J. C. Ng & A. A. Scawright. 1994. The magical and medicinal usage of Stangeria eriopus in South Africa. Journal of Ethnopharmacology 43: 67-72.

Pearson, H. H. W. 1905. Notes on south African cycads. Transactions of the South African Philosophical Society 16: 341-354.

Pearson, H. H. W. 1916. The Kirstenbosch cycads. Journal of the Botanical Society 2: 7-13.

Perez-Farrera. M. A. & A. P. Vovides. 2006. The ceremonial use of the threatened "Espadana" cycad (Dioon merolae, Zamiaccae) by a community of the central depression of Chiapas, Mexico. Boletin de la Sociedad Botanica de Mexico 78: 107 113.

Perez-Farrera, M. A., A. P. Vovides, P. Octavio-Aguilar, J. Gonzalez-Astorga, J. de la Cruz-Rodriguez, R. Hernandez-Jonapa, S. M. Villalobos-Mendez & M. A. Perez-Farrerra. 2006. Demography of the cycad Ceratozamia mirandae (Zamiaceae) under disturbed and undisturbed conditions in a biosphere Reserve of Mexico. Plant Ecology 187(1): 97-108.

Phelan, J., H. Van Hensbergen & R. Osborne. 1993. The apparent seed viability of Encephalartos ferox growing in sun and shade conditions in the wild. Encephalartos 34: 11-14.

Prado, A. 2011. The cycad herbivores. Bulletin de la Societe d'entomologie du Quebec. Antennae 18(1): 3-6.

Prakash, S. & J. Van Staden. 2008. Genetic variability and species identification within Encephalartos using random amplified polymorphic DNA (RAPD) markers. South African Journal of Botany 74: 735-739.

Proches, & S. D. Johnson. 2009. Beetle pollination of the fruit-scented cones of the south African cycad Stangeria eriopus. American Journal of Botany 96: 1722-1730.

Radha, P. & R. Singh. 2008. Ethnobotany and conservation status of Indian Cycas species. Encephalartos 93(1): 15-21.

Raimondo, D. 2001. Investigating the impacts of plant collecting on the population dynamics of two cycad spccics using population projection matrices and elasticity analyses. Masters dissertation. University of Cape Town.

Raimondo, D. & J. S. Donaldson. 2003. Responses of cycads with different life histories to the impact of plant collecting: Simulation models to determine important life history stages and population recovery times. Biological Conservation 111: 345-358.

Raimondo, D., L. Von Staden, W. Foden, J. E. Victor, N, A. Helme, R. C. Turner, D. A. Kamundi & P. A. Manyama (eds). 2009. Red list of south African plants. Strelitzia 25. South African National Biodiversity Institute, Pretoria.

Rattray. G. 1913. Notes on the pollination of some south African cycads. Transactions of the Royal Society of South Africa 3, 259-270.

Ravele, A. M. & R. A. Makhado. 2008. The value of cycads to human wellbeing--An overview. Encephalartos 93: 23-26.

Ravele, A. M. & R. A. Makhado. 2009. Exploitation of Encephalartos transvenosus outside and inside Mphaphuli cycads nature reserve, Limpopo Province, South Africa. African Journal of Ecology 48: 105-110.

Rayner. T. & A. Pires. 2016. Protecting highly threatened south African cycads from extinction: A new approach. Veld & Flora 102(1): 14-16.

Retief, K., A. G. West & M. F. Pfab. 2014. Can stable isotopes and radiocarbon dating provide a forensic solution for curbing illegal harvesting of threatened cycads? Journal of Forensic Sciences 59(6): 1541-1551.

Retief, K., A. G. West & M. F. Pfab. 2015. Arc you involved in the illegal cycad trade? Public misconceptions which arc detrimental to the survival of South Africa's cycads. Veld & Flora 101(1): 13-15.

Robbertse. H.. T. Naudc & R. Rousseau. 1999. Encephalartos middelburgensis population at Rhenosterpoort. Encephalartos 58: 26-29.

Rousseau, P. 2011. Systematic analysis of the African endemic cycad genera Encephalartos Lehm. And Stangeria T. Moore. MSc. Dissertation. University of Johannesburg.

Rousseau, P. 2013. Successive male cone production in Encephalartos. Encephalartos 114: 26-32.

Rousseau, R. & P. Rousseau. 2011. Encephalartos middelburgensis on site propagation project: Progress and prospects with lessons learnt. Enccphalartos 104: 20-24.

Sass, C., D. P. Little. D. W. Stevenson & C. D. Specht. 2007. DNA barcoding in the Cycadales: Testing the potential of proposed barcoding markers for species identification of cycads. PLoS One 2 (11): cl 154.

Schneider. D.. M. Wink. F. Sporer & P. Lounibos. 2002. Cycads: Their evolution, toxins, herbivores and insect pollinators. Naturwissenschaften 89: 281-294.

Scott-Shaw, C. R. 1995. Demographic studies of Encephalartos ghellinckii. Pp 435-439. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. The Cycad Society of South Africa. Stellenbosch.

Scott-Shaw, C. R. 1999. Rare and threatened plants of KwaZulu-Natal and Neighbouring regions. KwaZulu-Natal Nature Conservation Service, Congella. South Africa.

Silvertown, J. W. 1980. The evolutionary ecology of mast seeding in trees. Biological Journal of the Linnean Society 14: 235-250.

Singh, K. 2012. Population dynamics of the Zuurberg cycad and the predicted impact of climate change. Masters Dissertation. Nelson Mandela Metropolitan University. Port Elizabeth.

Snow, E. L. & G. H. Walter. 2007. Large seeds, extinct vectors and contemporary ecology: Testing dispersal in a locally distributed cycad, Macrozamia lucida (Cycadalcs). Australian Journal of Botany 55: 592-600.

Staude, H. 2001. African cycads and moths: An intricate relationship of ancient origin. Pp 307-311. In: D. G. Goode. Cycads of Africa, Volume I. D & E Cycads of Africa Publishers. Sandten.

Stiles, D. 1981. The Boni: Problems of a hunting-gathering people. Africana 8(2): 23-25.

Stobart. M. 1989. Beetles and pollination in Encephalartos altensteinii. Enccphalartos 17: 32.

Suinyuy, T. N., J. S. Donaldson & S. D. Johnson. 2009. Insect pollination in the African cycad Encephalartos friderici-guilielmi Lehm. South African Journal of Botany 75: 682-688.

Suinyuy, T. N., J. S. Donaldson & S. D. Johnson. 2010. Scent chemistry and patterns of thermogenesis in male and female concs of the African cycad Encephalartos natalensis (Zamiaceac). South African Journal of Botany 76: 717-725.

Suinyuy, T. N., J. S. Donaldson & S. D. Johnson. 2012. Geographical variation in cone volatile composition among populations of African cycad Encephalartos villosus. Biological Journal of the Linnean Society 106: 514-527.

Suinyuy, T. N., J. S. Donaldson & S. D. Johnson. 2013a. Patterns of odour emission, thermogenesis and pollinator activity in cones of an African cycad: What mechanisms apply? Annals of Botany 112: 891-902.

Suinyuy, T. N., J. S. Donaldson & S. D. Johnson. 2013b. Variation in the chemical composition of cone volatiles within the African cycad genus Encephalartos. Phytochemistry 85: 82-91.

Tang, W. 1987. Heat production in cycad cones. Botanical Gazette 148: 165-174.

Tang, W., L. Sternberg & D. Price. 1987. Metabolic aspects of thermogenesis in male cones of five cycad species. American Journal of Botany 74: 1555-1559.

Teny, I., C. J. Moore, G. H. Walter, P. I. Forster, R. B. Rocmer, J. S. Donaldson & P. J. Machin. 2004. Association of cone thermogenesis and volatiles with pollinator specificity in Macrozamia cycads. Plant Systcmatics and Evolution 243: 233-247.

Terry, I., G. H. Walter, C. J. Moore. R. B. Rocmer & C. Hull. 2007. Odor-mediated push-pull pollination in cycads. Science 318(5847): 70.

Thieret, J. W. 1958. Economic botany of the cycads. Economic Botany 12(1): 3-41.

Van der Walt, K. 2010. The critical difference between extinction and survival: Ex situ conservation of Encephalartos species in the Lowveld National Botanical Garden, South Africa. Enccphalartos 100: 1116.

Varghese, A. & T. Ticktin. 2006. Harvest, trade, and conservation of the endemic multiuse cycad, Cycas circinalis L., in the Nilgiri Biosphere Reserve, South India. People and Plants International and Keystone Foundation. Unpublished report, http://www.docstoc.com/docs/83695331/ An-assessment-of-theimpactsof-harvcst-on-Cycas-circinalis.

Vice, A. 1995. Encephalartos altensteiniv. A massive rape but possible recovery. Encephalartos 44: 15-19.

Vorster, P. & E. Vorster. 1985. Focus on Stangeria eriopus. Enccphalartos 198(5): 8-17.

Vorster, P. 2004. Growth form and habitat preference of the Cycadales. Pp 121-130. In: J.A. Lindstrom (ed). The biology, structure and Systematics of the Cycadales, Proceedings of the Sixth International Conference on Cycad Biology, Nong Nooch Tropical Botanical Garden, Thailand.

Vovidcs, A., N. Ogato & V. Sosa. 1997. Pollination of endangered Cuban cycad Microcycas calocoma (Miq.) a.DC. Botanical Journal of the Linnean Society 125: 201-210.

Walter, K. S. & H. J. Gillett (eds). 1998. IUCN red list of threatened plants. Compiled by the world conservation monitoring Centre. IUCN -The World Conservation Union. Gland, Switzerland and Cambridge, UK

Walters, T. 2003. Off-site collections. Pp 48-53. In: J. S. Donaldson (cd). Cycads: A status survey and conservation action plan. IUCN/SSC Cycad Specialist Group, Gland.

Watt, J. M. & M. G. Breyer-Brandwijk. 1962. The medical and poisonous plants of southern and eastern Africa, 2nd edition. Livingstone.

Whitelock, L. M. 2002. The Cycads. Timber Press Inc. Portland, Oregon.

Williams, V. L. 2003. Hawkers of Health: An investigation of the Faraday street traditional medicine market in Johannesburg, Gauteng. Plant ecology and conservation series no. 15 (report to Gauteng Directorate of Nature Conservation. DACEL). University of the Witwatersrand, Johannesburg.

Williams, V. L., K. Balkwill & E. T. F. Witkowski. 2001. A lexicon of plants traded in the Witwatersrand umuthi shops, South Africa. Bothalia 31:71-98.

Williams, V. L., S. R. Cousins & E. T. F. Witkowski. 2014. From fragments to figures: Estimating the number of Encephalartos stems in a muthi market. South African Journal of Botany 93: 242-246.

Williams, V. L., E. T. F. Witkowski & K. Balkwill. 2007. Volume and financial value of species traded in the medicinal plant markets of Gauteng, South Africa. International Journal of Sustainable Development and World Ecology 14: 584-603.

Williamson, ,l., O. Maurin. S. Shiba, Van der Bank, H.. M. Pfab, M. Pilusa, R. Kabongo & Van der Bank, M.. 2016. Exposing the illegal trade in cycad species (Cycadophyta: Encephalartos) at two traditional medicine markets in South Africa using DNA barcoding. Genome 59(9): 771-781.

Woodcnbcrg, W., D. P. Etdey, N. W. Pammenter & P. Bcijak. 2007. Post-shedding seed behaviour of selected Encephalartos species. South African Journal of Botany 73(3): 496.

Zobolo, A. M. & Q. N. Mkabela. 2006. Traditional knowledge transfer of activities practised by Zulu women to manage medicinal and food plant gardens. African Journal of Range & Forage Science 23(1): 77-80.

Zunckel. K. 1990. The ecology and management of the Kaapschoop cycad (Encephalartos laevifolius Stapf and Burtt Davy). PhD thesis. University of Cape Town.

Zunckel. K. 1995. The role of insects and fire in the ecology of Encephalartos laevifolius and their management implications. Pp 287-293. In: P. Vorster (ed). Proceedings of the third international Conference on cycad biology. Cycad Society of South Africa, Stellenbosch.

S. R. Cousins (1,2) * E. T. F. Witkowski (1)

(1) School of Animal, Plant and Environmental Sciences, University of the Witwatersrand, Private Bag 3, Wits, Johannesburg 2050, South Africa

(2) Author for Correspondence; e-mail: swartland.botanics@gmail.com

Published online: 1 May 2017

Caption: Fig. 1 Geographical distributions of the three cycad genera on the African continent: (a) Encephalartos (65 species), (b) Cycas thouarsii and (c) Stangeria eriopus. Countries shaded in grey contain at least one cycad species. Shaded squares denote Degree Squares (DS)

Caption: Fig. 2 The three cycad genera that occur on the African continent, (a-d) Four representatives of the genus Encephalartos, showing variation in growth form, leaf size and colour, (a) Encephalartos latifrons--a large, arborescent, broad-leaved species endemic to the Eastern Cape. South Africa (b) Encephalartos cycadifolius a semi-subterranean-stemmed, fine-leaved grassland species endemic to the Eastern Cape, South Africa (e) A subterranean-stemmed species--Encephalartos villosus--endemic to South Africa and Swaziland (d) Encephalartos horridus--a dwarf arborescent, so-called "blue leaf' species endemic to the Eastern Cape, South Africa (e) Stangeria eriopus--a subterranean-stemmed cycad endemic to grasslands and forests of South Africa and southern Mozambique (f) Cycas thouarsii--the only representative of the genus Cycas on the African continent. Photo credits: a, b & c (John Donaldson), d (Gerhard van Devcnter), e (David Muller), f (Stephen Cousins)

Caption: Fig. 3 (a) Coralloid roots at the base of an Encephalartos arenarius stem, just above the soil surface (b) Encephalartos seeds germinating on the soil surface beneath the mother plant in habitat. Some of the seeds still show the remains of their desiccating sarcotestae (c) Apex of a coning Encephalartos friderici-guilielml plant showing emerging cataphylls around the base of each cone, as well as a layer of tomcntum or "wool" between the older cataphylls lower down (d) The damaged stem of an Encephalartos transvenosus individual from which bark has been harvested for traditional medicine. Photo credits: a & c (John Donaldson), b (Gerhard van Deventer), d (Xander de Kock)

Caption: Fig. 4 Cone morphology of two representatives of the genus Encephalartos, Stangeria eriopus and Cycas thouarsii. The tomentose cones of (a) a male and (b) a female Encephalartos friderici-guilielmi individual. This specics is one of the seven so-called "woolly-coned" Encephalartos species endemic to South Africa (c) Glabrous male and (d) female cones of Encephalartos ferox (e) Male and (f) female cone of Stangeria eriopus (g) Male and (h) female conc of Cycas thouarsii. Photo credits: a, b, e & f (John Donaldson), c & g (Michael Calonje), d (Piet Vorster), h (Stephen Cousins)

Caption: Fig. 5 Invertebrates that utilize Encephalartos species as a food source (a) Adult female Leopard magpie moth (Zerenopsis lepida) with eggs on a newly-emerging Encephalartos septentrionalis leaf (b) Leopard magpie moth larva on Encephalartos friderici-guilielmi (c) The weevil Antliarhinus zamiae, which feeds exclusively on cycad seeds (scale bar = ~5 mm) (d) A male Callioratus abraxus moth. Photo credits: (a & b) Gerhard van Deventer, (c & d) John Donaldson

Caption: Fig. 6 Number of cycad spccies (Cycas thouarsii, Encephalartos species and Stangeria eriopus) and their Red List statuses (IUCN, 2016) in each African country in which cycads arc known to occur. EW = Extinct in the Wild. CR = Critically Endangered, EN = Endangered. VU = Vulnerable, NT = Near Threatened, and LC = Least Concern
Table 1 Summary of studies conducted on African cycad species
documenting cone visitation and/or pollination by insects

Cycad species           Cone visitors           Role in pollination

Encephalartos           Antliarhinus sp.        Pollen-bearing
altensteinii                                    individuals observed
                                                in both male and
                                                female cones

                        Nine beetle species     Four species occurred
                        (names not stated),     regularly on cones of
                        eight of which were     both sexes; evidence
                        curculionids            of pollen transfer
                                                from male to female
                                                cones for two species

Encephalartos           Metacucujus             Major pollinator
cycadifolius            encephalarti
                        (Coleoptera;
                        Cucujoidca) and an
                        Erotylidae sp.

Encephalartos           Platymerus sp.          Visitor to male and
friderici-guilielmi     (weevil)                female cones

                        Porthetes hispidus      Present in sufficient
                        (Curculionidae),        numbers during
                        Metacucujus             pollination to be
                        encephalarti            potential pollinators
                        (Cucujoidca) and an
                        undcscribed
                        Erotylidae sp.

Encephalartos           Porthetes sp.           Visitor to male cones
laevifolius             (Coleoptera;
                        Curculionidae)

Encephalartos           Antliarhinus            Antliarhinus signatus
natalensis              signatus, A. zamiae.    and A. zamiae were
                        and three beetle        only observed on the
                        species: Metacucujus    surface of female
                        goodei (Boganiidae),    cones prior to
                        an undcscribed          receptivity. The
                        Porthetes sp.           beetle spccies were
                        (Curculionidae), and    present in receptive
                        an Erotylidae sp.       female cones and in
                        nov.                    male cones during
                                                pollen shed, except
                                                for Metacucujus
                                                goodei, which was not
                                                observed visiting
                                                female cones

Encephalartos           Metacucujus             Visitor to male cones
transvenosus            encephalarti            only; no insects
                                                observed in female
                                                cones

Encephalartos           Porthetes sp.           Major pollinator
villosus                (Coleoptera:
                        Curculionidae)

                        Antliarhinus zamiae     Minor pollinator

                        Undescribed beetle      Minor pollinator
                        species
                        (Xenoscelinae;
                        Languriidac)

                        Metacucujus goodei      Minor to no role
                        (Boganiidae)

                        Unidentified beetle     Minor to no role
                        species
                        (Xenoscelinae)

                        Porthetes sp.           Visitor to male and
                        (Coleoptera;            female cones
                        Curculionidae),
                        Erotylidae sp. nov.,
                        Metacucujus goodei
                        (Boganiidae)
                        Antliarhinus zamiae

Stangeria eriopus       Mostly sap and rove     Of all the visitors,
                        beetles (Coleoptera:    only sap beetles
                        Nitidulidae,            (Nitidulidae) were
                        Staphylinidae) and      able to effect
                        fruit flies (Diptera:   pollination under
                        Drosophilidae).         experimental
                                                conditions

Cycad species           Cone visitors           Reference

Encephalartos           Antliarhinus sp.        Rattray (1913)
altensteinii

                        Nine beetle species     Stobart (1989)
                        (names not stated),
                        eight of which were
                        curculionids

Encephalartos           Metacucujus             Donaldson et al.
cycadifolius            encephalarti            (1995)
                        (Coleoptera;
                        Cucujoidca) and an
                        Erotylidae sp.

Encephalartos           Platymerus sp.          Oberprieler (1989)
friderici-guilielmi     (weevil)

                        Porthetes hispidus      Suinyuy ct al. (2009)
                        (Curculionidae),
                        Metacucujus
                        encephalarti
                        (Cucujoidca) and an
                        undcscribed
                        Erotylidae sp.

Encephalartos           Porthetes sp.           Zunckel (1995)
laevifolius             (Coleoptera;
                        Curculionidae)

Encephalartos           Antliarhinus            Suinyuy et al. (2010)
natalensis              signatus, A. zamiae.
                        and three beetle
                        species: Metacucujus
                        goodei (Boganiidae),
                        an undcscribed
                        Porthetes sp.
                        (Curculionidae), and
                        an Erotylidae sp.
                        nov.

Encephalartos           Metacucujus             Grobbelaar (1999)
transvenosus            encephalarti

Encephalartos           Porthetes sp.           Donaldson (1997)
villosus                (Coleoptera:
                        Curculionidae)

                        Antliarhinus zamiae     Donaldson (1997)

                        Undescribed beetle      Donaldson (1997)
                        species
                        (Xenoscelinae;
                        Languriidac)

                        Metacucujus goodei      Donaldson (1997)
                        (Boganiidae)

                        Unidentified beetle     Donaldson (1997)
                        species
                        (Xenoscelinae)

                        Porthetes sp.           Suinyuy et al. (2012)
                        (Coleoptera;
                        Curculionidae),
                        Erotylidae sp. nov.,
                        Metacucujus goodei
                        (Boganiidae)
                        Antliarhinus zamiae

Stangeria eriopus       Mostly sap and rove     Proches and Johnson
                        beetles (Coleoptera:    (2009)
                        Nitidulidae,
                        Staphylinidae) and
                        fruit flies (Diptera:
                        Drosophilidae).
COPYRIGHT 2017 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cousins, S.R.; Witkowski, E.T.F.
Publication:The Botanical Review
Article Type:Report
Geographic Code:6KENY
Date:Jun 1, 2017
Words:21741
Previous Article:Herbarium-based preliminary conservation assessments of Lesser Antillean endemic seed plants reveal a flora at risk.
Next Article:Fruit and seed dispersal of Salvia L. (Lamiaceae): a review of the evidence.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |