Epiphyte diversity on host trees in an urban environment, eThekwini Municipal Area, South Africa

Abstract

Vascular epiphytes live non-parasitically on other plants and are a distinctive and intergral component of tropical forests. There is a general lack of studies examining epiphyte diversity in urban settings. The aim of this study was to document the diversity of epiphytes on host trees in the eThekwini Metropolitan Area (EMA). In addition, the number of individuals of each epiphyte, host tree circumference at breast height and height were recorded. In total, 30 epiphyte species from 12 families were recorded, with most epiphyte species belonging to the Moraceae (n = 8) and Araliaceae (n = 5). A total of 34 host species from 15 families were recorded. These numbers did not increase when considering herbarium material within the EMA sensu stricto. The highest epiphyte richness (n = 13) was reported on the palm Raphia australis. The high number of both alien host and epiphytic taxa are of concern. More studies are needed to understand how epiphyte/host interactions in the urban landscape are established and maintained.

Keywords:

• diversity
• epiphyte
• eThekwini Municipal Area
• South Africa
• urban landscape

Introduction

Epiphytes are plants that grow on other plants for structural support and anchorage, and not for water or nutrient supplies (i.e. non-parasitic) (Laube & Zotz 2006). Approximately 84 vascular plant families have evolved an epiphytic strategy and at least 876 genera have one epiphytic species (Gentry & Dodson 1987; Kress 1989). The number of epiphytes is estimated to be c. 29,000 species (Gentry & Dodson 1987). Epiphytes are considered an important component of global plant diversity and represent c. 10% of the world’s vascular plant species (Gentry & Dodson 1987; Nieder et al. 2001).

Araceae, Bromeliaceae, Orchidaceae and several fern families have a substantial number of epiphytes (Fernandez Monteiro et al. 2009). Epiphytes can be broadly classified as holoepiphytes (spending their entire life cycle in the canopy) and hemiepiphytes (spending some stage of their life rooted in terrestrial soil). Some epiphytes start life in the canopy and send roots to the ground (primary hemiepiphytes), whereas others start on the ground, grow to the canopy and lose terrestrial connections (secondary hemiepiphytes) (Benzing 2004; Lowman & Rinker 2004). Epiphytes are ecologically important, playing a significant role in forest ecosystem processes and interactions, and contribute to species richness (Cummings et al. 2006; Burns & Zotz 2010). Epiphytes are considered to be indicators of ecosystem diversity, health and productivity (Jovan & McCune 2006; Bartels & Chen 2012). They provide nutrients for other flora and fauna, and are also very sensitive to shifts in microclimate (Gradstein 2008). Epiphytes play an important role in tropical ecosystem functioning (Madison 1977), by producing relatively high amounts of organic matter on host trees and fixing atmospheric nitrogen (Nadkarni 1984). Epiphytes are also considered an integral part of some urban ecosystems and constitute an important structural and functional component (Ellis & Coppins 2007). Epiphytic plants provide water, food (nectar, fruits), habitat and nesting materials for invertebrates and birds (Affeld 2008; Alvarenga et al. 2010). Díaz et al. (2012) have shown that trees with epiphytes have greater invertebrate species richness and abundance compared with trees with epiphytes removed.

Epiphyte distributions are influenced by stand characteristics, such as stand age and tree species composition, as well as by dispersal limitation (Woods 2013). A host tree species may contain a diverse community of epiphytes that is often different from other host tree species (Laube & Zotz 2006). Host-specific differences in epiphyte assemblages suggest that epiphyte diversity may be related to variation in microhabitats within individual host trees (Cardelús & Chazdon 2005). Some authors have reported that epiphyte growth on particular host trees is also affected by several physical and chemical host traits, and environmental factors (Vanderpoorten et al. 2004). The preference of a specific epiphyte species for a particular host tree species has been reported repeatedly (Benzing 1990; Male & Roberts 2005; Laube & Zotz 2006). Several studies have examined epiphytic flora on single host species (Affeld et al. 2008; Werner & Gradstein 2009; Larreaa & Werner 2010). A total of 126 morphospecies (52 genera and 21 plant families) were found growing epiphytically on a single mature host tree (phorophyte) in Costa Rica (Schuettpelz & Trapnell 2006). Other studies have examined a limited number of phorophytes. Freiberg (1996) recorded a total of 77 species (17 families) from three phorophyte tree species in French Guiana.

Epiphytes contribute to species richness and play a substantial role in the processes and interactions that make a forest function (Cummings et al. 2006). They constitute a large proportion of photosynthetically active material (Hofstede et al. 2001), and significantly influence important processes, particularly mineral and hydrological cycles (Benzing 1998). Besides that, epiphytes play key roles in forest dynamic processes and they are a major contributor to local, regional and global plant diversity (Batke 2012).

Urban green spaces (parks, avenues, greenways, etc.) provide an important contribution to quality of life by providing habitats for a range of species (Bullock 2008; Savard et al. 2000). Ecological studies on epiphytes have been conducted in several urban settings across the globe (Fudali 2007, 2012; Richter et al. 2009; Adhikari et al. 2012). However, limited studies have examined epiphytes in South Africa (Zotz 2005), and no studies have examined epiphytic plants in an urban setting. Information on epiphytes is required in order to develop potential conservation and management strategies. Hence, this study was carried out to investigate diversity of epiphytes on host trees in the eThekwini Municipal Area (EMA), which will provide baseline information that could guide conservation and management of urban epiphyte communities.

Material and methods

Study area

The study was performed within the EMA from March to July 2012. Selected localities within the EMA, viz. Che Guevara Road (Moore), Clark Road, Helen Joseph Road, Florida Road, and Randles Road (Table 1), which have a notable presence of epiphytes, were investigated. The sites were based on the extensive field experience of one of the authors (YG). These sites have trees primarily planted and maintained by the eThekwini Municipality. Mean annual rainfall of the study area is between 1000 and 1250 mm. The mean annual temperature is 20.5°C with a variance of 8.3°C. The study region has a subtropical and humid climate, with rainfall mostly falling in the summer months. However, there is some form of precipitation in all seasons (Roberts 1993), which allows continuous plant growth throughout the year. The natural vegetation is a mosaic of coastal forest, savanna, grassland, thicket and mangroves (Roberts 1993). Municipal authorities manage trees along roadsides, parks and other green spaces (eThekwini Municipality 2007). It is important to note that there is no active planting of epiphytes on street or park trees within the municipality.

Table 1 Sites surveyed within the eThekwini Municipal Area (EMA) during March–July 2012.

Street name	GPS coordinates
Che Guevara Road	29°51′07″S 30°59′22″E
Clark Road	29°51′13″S 30°59′19″E
Helen Joseph Road	29°51′26″S 30°59′30″E
Florida Road	29°49′39″S 31°00′37″E
Randles Road	29°49′35″S 30°58′56″E

Data collection and analysis

Visual searches were made for all epiphyte species on trees along public roads included in this study. The field survey was carried out using a ladder. Binoculars were used to detect epiphyte individuals occurring on distant branches. The number of host trees was recorded. The growth habit of all epiphytes and hosts (phorophtyes) was documented. Phorophytes with fewer than 20 individual trees across the entire study site were recorded, but excluded from further analysis, because some of these were difficult to sample with hampered access to the canopy for examination of epiphytes. Consequently, only nine frequent host species comprising 254 trees were found to be appropriate for sampling (i.e. with unhampered canopy assess for examination of epiphytes) (Fig. 2). The circumference at breast height (CBH, i.e. 1.37 m from the ground), height, bark characteristics and number of epiphyte species on phorophytes were recorded. The height of each host tree was measured with a clinometer, or estimated visually. Three different rugosity classes (smooth, medium and rough) were delimited based on the roughness of the host tree bark. All taxa were identified using published accounts and with the aid of herbarium material at the Ward Herbarium (UDW) and KwaZulu-Natal Herbarium (NH). The epiphytic taxa were catagorised into their typical life forms as well as epiphytic type. Epiphytes were divided into two types following Kersten et al. (2009) namely specialist (found in only one species of host tree) and selective (frequently found on few species of host trees and rarely on other species).

Herbarium material and existing literature for all taxa surveyed (phorophytes and epiphytes of the present study) were checked in order to determine if epiphytes had a wider pool of host species in the study area. Records of epiphytism (both herbarium and literature) within and beyond the EMA were noted. These cases represent potential epiphyte–phrorphyte interactions of taxa known to occur within the EMA, but were not detected.

Results

Host taxa

Thirty-four phorophyte tree species from 15 families were observed. Twenty species were alien to South Africa, with 14 indigenous South African species (Appendix 1). Nine species were considered the most frequent phorophytes (i.e. housed more than 20 epiphytic individuals), and included Brachychiton acerifolium, Casuarina equisitifolia, Colvillea racemosa, Delonix regia, Jacaranda mimosifolia, Phoenix reclinata, Pleiogynium timorense, Raphia australis and Trichilia dregeana. The remaining 25 phorophyte species had a poor representation of epiphytes (fewer than 20 individuals).

Epiphytic taxa

Epiphyte species composition varied substantially among host species. A total of 30 epiphyte species from 12 families were recorded, with most epiphyte species belonging to the Moraceae (n = 8) and Araliaceae (n = 5). In terms of life forms, there were 17 trees, 2 shrubs and 11 herbs (forbs) for the epiphytic taxa. A total of 13 epiphytic species are found to be specialist (found on only one host species), whereas 17 are selective (epiphyte species found on more than one host species) (Appendix 1). Only 13 of the 30 epiphytic species observed were native to South Africa (Table 2, Appendix 1).

Table 2 List of the epiphyte species (n = 30) recorded during study.

Species	Family	Nativity	Normal life form	Epiphytic type/form
Ageratum houstonianum	Asteraceae	Mexico	Herb	Pseudoepiphyte
Amaranthus tricolor	Amaranthaceae	Asia	Herb	Pseudoepiphyte
Ansellia africana	Orchidaceae	Indigenous	Herb	Holoepiphyte
Arecastrum romanzoffianum	Arecaceae	Mexico	Tree	Pseudoepiphyte
Asystasia gangetica	Acanthaceae	India/Malaysia	Herb	Holoepiphyte
Cassia multijuga	Fabaceae	Brazil	Tree	Pseudoepiphyte
Chromolaena odorata	Asteraceae	America	Shrub	Pseudoepiphyte
Chrysophyllum oliviforme	Sapotaceae	Florida	Tree	Pseudoepiphyte
Cussonia spicata	Araliaceae	Indigenous	Tree	Pseudoepiphyte
Cyrtorchis arcuata	Orchidaceae	Indigenous	Herb	Holoepiphyte
Ficus benghalensis	Moraceae	India	Tree	Hemiepiphytes
F. bubu	Moraceae	Indigenous	Tree	Hemiepiphytes
F. burt-davyii	Moraceae	Indigenous	Tree	Hemiepiphytes
F. lutea	Moraceae	Indigenous	Tree	Hemiepiphytes
F. natalensis	Moraceae	Indigenous	Tree	Hemiepiphytes
F. polita	Moraceae	Indigenous	Tree	Hemiepiphytes
F. thonningii	Moraceae	Indigenous	Tree	Hemiepiphytes
Livistona chinensis	Arecaceae	South China	Tree	Pseudoepiphyte
Morus alba	Moraceae	China	Tree	Pseudoepiphyte
Mystacidium venosum	Orchidaceae	Indigenous	Herb	Holoepiphyte
Platycerium bifurcatum	Polypodiaceae	Indigenous	Herb	Holoepiphyte
Pleopeltis macrocarpa	Polypodiaceae	Indigenous	Herb	Holoepiphyte
Polystichum munitum	Dryopteridaceae	Japan	Herb	Holoepiphyte
Rumohra adiantiformis	Dryopteridaceae	Indigenous	Herb	Holoepiphyte
Schefflera actinophylla	Araliaceae	Australia	Tree	Hemiepiphytes
Schefflera arboricola	Araliaceae	Subtropical regions	Shrub	Hemiepiphytes
Schefflera umbellifera	Araliaceae	Australia	Tree	Hemiepiphytes
Schefflera venulosa	Araliaceae	Australia	Tree	Pseudoepiphyte
Solanum mauritianum	Solanaceae	Brazil	Tree	Pseudoepiphyte
Stenochlaena tenuifolia	Polypodiaceae	India	Herb	Holoepiphyte

In most cases, herbarium label data of both host and epiphyte taxa (recorded in this study) did not specify either host or epiphyte details. When combined with the taxa of the present study, the herbarium and literature survey (Appendix 2) revealed 52 phorophytes and 33 epiphytes.

In an effort to facilate reporting, major epiphytic taxa are grouped based on phylogenetic affinities: pteridophytes (ferns and allies), Moraceae (viz. Ficus), Araliaceae (viz. Schefflera) and Orchidaceae. Ferns are well known for epiphytism (Roux 2001). Stenochlaena tenuifolia is known to be epiphytic when mature (Roux 2001; Crouch et al. 2011) (see Appendix 2 for herbarium and literature records).The genus Platycerium is known to be epiphytic (Kreier & Schneider 2006). Platycerium bifurcatum is a mid- to high-level epiphyte sometimes observed on concrete bridges (Roux 2001; Crouch et al. 2011). It has escaped from cultivation and is naturalised in the greater Durban region (Crouch et al. 2011). Polystichum is known to be epilithic or epiphytic (Roux 2001). Polystichum minitum has been recorded as epiphytic (Williams & Sillett 2007). Pleopeltis macrocarpa is either epilithic or epiphytic (Roux 2001), and regarded by Crouch et al. (2011) as one of the most widespread epiphytes in southern Africa. Rumohra adiantiformis is recorded as being terrestrial, epilithic or low level epiphytic (Roux 2001).

The genus Ficus (Moraceae) is well documented to be epiphytic (Burrows & Burrows 2003). Several South African figs are known to have a strangling habit. Representatives of the Ficus thonningii complex (F. rokko, F. psilopoga, F. persicifolia, F. petersii, F. burkei and F. natalensis) have a strangler/hemiepiphytic habit. Ficus lutea, F. polita and F. natalensis are hemiepiphytic or epiphytic at first before becoming free-standing trees (Burrows & Burrows 2003). Ficus burtt-davyi is known to be an occasional hemiepiphyte or strangler. Ficus bubu can be hemiepiphytic. Ficus benghalensis is not known to be epiphytic (Burrows & Burrows 2003). It is interesting to note that F. benghalensis can potentially be a host and epiphyte (Appendix 1). A detailed record of epiphytic figs from the herbarium records are provided in Appendix 2.

No herbarium records were found to suggest that Schefflera taxa are epiphytic within the EMA. However, Schefflera actinophylla has been observed to become a strangler in Durban. Several Schefflera taxa (S. abyssinica, S. goetzenii, S. myriantha) have an epiphytic habit (Cannon 1978). Schefflera actinophylla is a canopy epiphyte in the lowland rainforests of Viti Levu. This species can start out as an epiphyte but turn into a strangler (Keppel & Watling 2011). Both S. actinophylla and S. arboricola are sometimes hemiepiphytes (Tomlinson et al. 2005). Schefflera venulosa is known to be a climber (Wyatt-Smith 1953).

There are numerous epiphytic Orchidaceae species in South African including the EMA (Harrison 1981). Ansellia africana, Cyrtorchis arcuata and Mystacidium venosum were found in the study area. All three species are known to be epiphytic (Harrison 1981; Linder & Kurzweil 1999; McMurtry et al. 2008). No herbarium records were found within the EMA, however, several herbarium records were found for KwaZulu-Natal (Appendix 2). A summary of additional phorophytes of the of these orchids obtained from McMurtry et al. (2008) is given in Appendix 2.

Literature checks were carried out on all the remaining epiphytes (both genera and species) found in this study, and are summarised below. Ageratum houstonianum is not epiphytic; however, Ageratum conyzoides is known to be epiphytic (Akinsoji 1990). Asystasia gangetica is reported to be an epiphyte (Akinsoji 1990). Chromolaena odorata has been described as a diffuse scrambling shrub (Hilliard 1977), and is reported to be an epiphyte (Akinsoji 1990). The following taxa (n = 11) recorded in this study were not previously known to be epiphytic: Ageratum houstonianum, Amaranthus tricolor, Arecastrum romanzoffiana, Cassia multijuga, Chrysophyllum oliviforme, Cussonia spicata, F. benghalensis, Livistona chinensis, Morus alba, S. venulosa and Solanum  mauritianum.

The specificity analysis among epiphyte and host trees revealed that 13 epiphytic species are specialist (epiphyte species found on only one host species), whereas 17 are selective (epiphyte species found on more than one host species) (Appendix 1). In our study area, most of the epiphyte species seem to grow on tree forks of the host tree (Fig. 1). Epiphyte diversity on host trees with different CBH and height classes are presented in Fig. 2. In general, most epiphytes were recorded on trees with a CBH of 61–180 cm (n = 92) and heights of 11–20 m (n = 74). ANCOVA showed no significant relationships between epiphyte diversity and tree CBH (P > 0.05) and height (P > 0.05). This suggests that host tree CBH and height are not suitable predictors of epiphyte diversity. However, the epiphyte diversity showed a significant relationship with host tree species (P < 0.05) (Fig. 2).

Figure 1 Examples of South African epiphytes. A, Ficus lutea on Elaeis guineensis; B, Schefflera actinophylla on Delonix regia; C, F. thonningii on Caryota mitis; D, Schefflera actinophylla on Colvillea racemosa; E, F. thonningii on Halleria lucida.

Figure 2 A, Number of epiphytes on trees of different height class (m). (A: 6–10; B: 11–15; C: 16–20; D: >20); B, Number of epiphyte on trees of different CBH class. JM, Jacaranda mimosifolia; DR, Delonix regia; TD, Trichilia dregeana; BA, Brachychiton acerifolium; PT, Pleiogynium timorense; CR, Colvillea racemosa; CE, Casuarina equisetifolia; RA, Raphia australis; PR, Phoenix reclinata.

Discussion

The present study was conducted in an urban South African setting. Elsewhere, studies in urban settings mainly focused on epiphytic bryophytes and lichens (Fudali 2007, 2012; Richter et al. 2009). In total 34 host species were found, with 20 species being non-native to South Africa. Seventeen of the 30 epiphytes recorded were non-native to South Africa. The epiphytes associated with the highest number of phorophytes were F. natalensis, S. actinophylla, F. thonningii, F. lutea and Platycerium bifurcatum. Schefflera actinophylla and Platycerium bifurcatum are alien epiphytes.

The high number of alien epiphytes and phoropytes is concerning. Platycerium bifurcatum has become naturalised in Florida but is unlikely to become invasive there (Pemberton 2003). Schefflera actinophylla is of more concern. This species can start out as an epiphyte but turn into a strangler. There has been an expansion of this species in some urban areas of Fiji, and it may become problematic (Keppel & Watling 2011). Keppel & Watling (2011) have also noted the concerning spread of Spathodea campanulata in Fiji, a species also found in the current study. Both native and non-native host species have almost the same number of epiphytes, including native and alien epiphytes. Our findings suggest that urban settings are promoting facilitative interactions of both alien phorophytes and epiphytes. These preliminary findings are indicative of a scenario supporting invasional meltdown (Simberloff & von Holle 1999) in phorophytes and epiphytes. At present, it is difficult to test this hypothesis further without additional sampling and rigous testing. Host–epiphytes interactions are poorly documented for native and alien South Africa taxa. Therefore, such data will increased the understanding of the role of epiphytism in biological invasions.

Epiphytes depend on their host tree mainly for support, whereby individual host trees provide the substrate for the establishment of epiphyte species. The successful establishment of epiphytes on their hosts depends on several host tree traits such as size, age, branch quality and bark texture (Lie et al. 2009; Bartels & Chen 2012). Brown (1990) found that variations in microclimate and substrate diameter are the main factors responsible for epiphyte distribution patterns. Tree size and bark characteristics have been suggested to explain the presence of epiphytes on trees (Benzing 1990; Nieder et al. 2001; Cascante-Marin et al. 2009). It is generally expected that large adult trees should harbour more epiphytes than small hosts (Lopez-Villalobos et al. 2008). These results are based on studies conducted in different natural forest types and remnant forest patches. The ANCOVA results for epiphyte diversity and host tree characteristics (viz. tree CBH and height) did not show such relationships in the present study. Our results for this aspect are best treated as inclonclusive, possibly due to limited sampling. Brown (1990) found that variations in microclimate and substrate diameter are the main factors responsible for epiphyte distribution patterns. It is generally expected that large adult trees should harbour more epiphytes than small hosts (Lopez-Villalobos et al. 2008). Hence, host attributes like CBH and height did not appear important determinants or predictors of epiphyte diversity in urban settings.

The greatest epiphyte richness in this study was found on the palm R. australis, which has smooth bark. In general, smooth surfaces of host trunks prevent seeds lodging directly on the bark, as well as humus or water accumulation and retention (Wyse & Burns 2011). Akinsoji (1990) found that smooth-barked hosts (phorophytes) had poor epiphyte representation because they are unable to accumulate dust, debris and moisture for the germination and growth of the epiphytes. However, R. australis has persistent leaf stalks in a more or less rossete arrangement, that transverse almost the entire length of the main trunk. The trunk has numerous leaf joints that can trap debris and facilitate epiphyte establishment. This might explain why palms with a smooth bark host a high epiphyte richness. Thus, specific traits such as the architecture (e.g. R. australis) and microhabitat resource availability of hosts are presumed to be important in the establishment of resident epiphytes. The occurence and density of epiphytes are influenced by host characteristics (host bark, pH, water holding capacity of bark and bark rugosity), which play importent role in determining aggregation and abundance of epiphyte species (Adhikari 2013). It is also reported that some epiphyte species show a preference for particular host and host traits (Hirata et al. 2009). A more comprehensive study is required in wild host populations to confirm this assertion.

It is worth noting that figs are an important indigenous component of epiphytic diversity in the EMA. This is possibly due to the fact that many figs can have a scrambling habit making them capable of colonising a very wide variety of host taxa (Harrison et al. 2003). It seems probable the habitat preferences of hemiepiphytic figs in the present study reflects similar microsite conditions and requirements for germination and seedling establishment among host species (Harrison et al. 2003).

This study represents the first South African survey of epiphyte diversity in an urban setting, and recovered some interesting findings. There is a general paucity of data in the study area, perhaps because urban ecology is an emerging discipline in South Africa (Cilliers & Siebert 2012). More comprehensive documentation of epiphyte diversity on host trees within natural ranges is needed. This will help understand how epiphyte/host interactions in the urban landscape are established and maintained.

Acknowledgement

SG and AB thank UKZN for postdoctoral fellowships. YG thanks eThekwini Municipality PRC for support. HB and SR thank UKZN for research funding and support. Preshen Banwari is thanked for the map of the EMA.

Appendix 1 List of host and epiphyte species recorded during the study

Table

	Epiphytes
Host species	Ageratum houstonianum^ab	Amaranthus tricolor^b	Ansellia africana	Arecastrum romanzoffiana^b	Asystasia gangetica^b	Cassia multijuga^ab	Chromolaena odorata^b	Chrysophyllum oliviforme^ab	Cussonia spicata^a	Cyrtorchis arcuata	Ficus benghalensis^b	F. bubu	F. burtt-davyii^ab	F. lutea	F. natalensis	F. polita	F. thonningii	Livistona chinensis^ab	Morus alba^ab	Mystacidium venosum^ab	Platycerium bifurcatum	Pleopeltis macrocarpa^ab	Polystichum munitum^ab	Rumohra adiantiformis^ab	Schefflera actinophylla^b	S. arboricola^b	S. umbellifera^ab	S. venulosa^b	Solanum mauritianum^b	Stenochlaena tenuifolia^ab
Adansonia digitata	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Albizia adianthifolia	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	+	−	−	−	−	−
Brachychiton acerifolium^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	+	−	−	−	−
Butia capitata^c	−	+	−	−	−	−	+	−	−	−	+	−	−	+	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	+
Caryota mitis^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−
Causarina equisitifolia^c	−	−	−	−	−	−	−	−	−	−	−	+	−	−	+	−	+	−	−	−	−	+	−	−	+	−	−	−	−	−
Celtis africana	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	+	−	−	−	−	−	−	−	+	+	−	+	−	−
Coccothrinax argentata^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Colvillea racemosa^c	−	−	−	−	−	−	−	−	−	−	−	+	−	−	+	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−
Combretum molle	−	−	+	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Delonix regia^c	−	−	+	−	−	−	−	−	−	−	−	−	−	+	+	+	+	−	−	−	−	−	−	−	−	+	−	+	−	−
Elaeis guineensis^c	−	−	−	−	+	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Ficus benghalensis^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	+	−	−	−	+	−	−	−	−
F. macrophylla^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−
F. natida^c	−	−	−	+	−	−	−	−	+	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Halleria lucida	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Hyophorbe lagenicaulis^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−
Hyphaene natalensis	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Jacaranda mimosifolia^c	−	−	−	−	−	−	−	−	−	−	−	−	−	+	+	+	+	−	−	−	+	−	−	−	+	−	+	−	−	−
Juniperus chinensis^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Kigelia africana	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	+	−	−	−	+	−	−	−	−	−
Mangifera indica^c	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−
Melia azedarach^c	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Mimusops obovata	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−
Phoenix reclinata	−	−	−	−	−		−	−	−	−	+	−	−	+	+	−	−	−	+	−	−	−	−	−	+	−	−	−	+	−
Pleiogynium timorense^c	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−	−	+	−	−	−	+	−	−	−	+	−	−	−	−	−
Raphia australis	+	+	−	−	+		−	+	−	−	+	−	+	+	+	+	−	−	−	−	−	−	+	+	+	−	−	−	+	−
Sabal minor^c	−	−	−	−	−		−	−	−	−	−	−	−	+	−	−	−	−	−	−	+	−	−	−	−	−	−	−	−	−
Searsia chirindensis	−	−	−	−	−		+	−	−	+	−	−	−	+	+	−	−	−	−	−	−	−	−	−	−	−	−	+	−	−
Spathodea campanulata^c	−	−	−	−	−		−	−	−	−	−	−	−	−	−	−	+	−	−	−	+	−	−	−	−	−	−	−	−	−
Strychnos decussata	−	−	−	−	−		−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Strychnos madagascariensis	−	−	−	−	−		−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Tababueaumbelata^c	−	−	−	+	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−	−
Trichilia dregeana	−	−	−	−	−		−	−	−	−	−	−	−	+	−	+	+	−	−	−	+	−	−	−	+	−	−	−	−	−
Total	1	2	3	2	2	1	3	1	1	4	3	2	1	8	16	6	10	2	1	1	8	1	1	1	11	4	1	3	2	1

^a Specialist epiphyte; ^balien epiphyte; ^cAlien host.

Appendix 2 Summary of epiphytism found in herbarium material and the literature for phorophytes and epiphytes of present study (records of epiphytism extending beyond the study area are included based on the availability of data)

Table

Epiphyte	Phorophyte taxon/taxa	Herbarium specimen details	Literature source	Locality EMA/beyond
Stenochlaena tenuifolia (Blechnaceae)	Raphia australis	Strey 8194, Ubombo (NH)	–	Beyond EMA
	Syzygium cordatum	Lawn 698, Eshowe (NH)	–	Beyond EMA
	Syzygium sp.	Buthelezi 502, Mtunzini (NH)		Beyond EMA
Ficus craterostoma (Moraceae)	Newtonia sp.	Ward 147, Ubombo (NH)	–	Beyond EMA
Ficus lutea (Moraceae)	Acacia sp.	Vahrmeyer & Tölken 958, Ingwavuma (NH)	–	Beyond EMA
Ficus natalensis (Moraceae)	Mimusops obovata	Morris 602, Shongweni Dam (NH)	–	EMA
	Podocarpus falcatus	Ward 3719, Manguzi Forest (NH)	–	Beyond EMA
	Strychnos decussata	Ward 10136, Umbilo Road (NH)	–	EMA
Ficus stuhlmannii (Moraceae)	Dialium schlechteri	Ward 2280, Tembe Elephant Park (NH)	–	Beyond EMA
Ficus thonningii (petersii) (Moraceae)	Cyathea dregei	Ward 11394, Mgayi Hill (NH)	–	Beyond EMA
	Phoenix reclinata	Strey 5252, Durban Botanic Garden (NH)	–	EMA
	Syzygium cordatum	Ward 11394, Mgayi Hill (NH)	–	Beyond EMA
Ficus tremula (Moraceae)	Mimusops caffra	MacDevette 883, Manzengwenya (NH)	–	Beyond EMA
Ansellia africana (Orchidaceae)	Acacia sieberiana var. woodii	–	McMurtry et al. (2008)	Beyond EMA
	Breonadia salicina	–	McMurtry et al. (2008)	Beyond EMA
	Combretum imberbe	–	McMurtry et al. (2008)	Beyond EMA
	Jacaranda mimosifolia	–	McMurtry et al. (2008)	Beyond EMA
	Peltophorum africanum	–	McMurtry et al. (2008)	Beyond EMA
	Spirostachys africana	–	McMurtry et al. (2008)	Beyond EMA
Cyrtorchis arcuata (Orchidaceae)	Syzygium cordatum	Ward 6725, Port Shepstone (UDW)	–	Beyond EMA
		Ward & Rajh 11700, Lake Nhlabane (UDW)	–	Beyond EMA
Mystacidium venosum (Orchidaceae)	Breonadia salicina		McMurtry et al. (2008)	Beyond EMA
	Cola microcarpa	Ward 7042, False Bay (UDW)		Beyond EMA
	Englerophytum magalismontanum		McMurtry et al. (2008)	Beyond EMA
	Maytenus acuminata		McMurtry et al. (2008)	Beyond EMA
	Podocarpus latifolius		McMurtry et al. (2008)	Beyond EMA
	Strychnos henningsii	Ward 7042, False Bay (UDW)		Beyond EMA
	Syzygium cordatum		McMurtry et al. (2008)	Beyond EMA

NH, KwaZulu-Natal Herbarium; UDW, Ward Herbarium.