The Megadrile Fauna (Annelida: Oligochaeta) of Queen Elizabeth Park, South Africa: Species Composition and Distribution within Different Vegetation Types

ABSTRACT A recent survey in selected habitat types in Queen Elizabeth Park (protected area for more than 50 years) resulted in the recording of 1127 specimens of nine earthworm species inhabiting the park, of which only one microchaetid Tritogenia howickiana is indigenous to South Africa. The other eight species are peregrine, widely introduced, and belong to four families: Megascolecidae (Amynthas aeruginosus, Amynthas corticis, Amynthas gracilis, Amynthas minimus and Amynthas rodericensis), Glossoscolecidae (Pontoscolex corethrurus), Lumbricidae (Octolasion lacteum) and Acanthodrilidae, Benhamiinae (Dichogaster saliens). The most abundant are megascolecids (883 specimens), dominating in all vegetation types. Although coexistence of exotic species with indigenous species is noted, a decline in the endemics and dominance by exotics was observed.


INTRODUCTION
It is widely known and accepted that earthworms are supportive environmental components and play an important role in the functioning of ecosystems (Lavelle et al. 1997;Bhadauria et al. 2000;Shuster et al. 2002), thus being regarded as soil eco sys tem engineers (Jouquet et al. 2006). The role of earthworms in soil eco systems and their sensitivity to disturbances (thus acting as bio-indicators) depends on the earth worm species diversity. Beneficial effects of earthworms on soils are well do cu men ted. According to Lavelle (1988) and Salome et al. (2011), food supply, soil texture, ve ge tation type and pH values play a ma jor role in governing the earthworm densities. The quality and amount of above and below ground litter produced by plants influence earthworm populations and this may be ob served in some forests and native grasslands (Campana et al. 2002;Whalen 2004). Due to their relationship to soil ecosystem func tion, earthworm populations de pict spatial and temporal heterogeneity in patches and their population structure may be influenced by a change in vegetation and soil cha rac teristics, as well as biotic and abi otic interactions (Margerie et al. 2001;Whalen 2004;Fey 2010;Valckx et al. 2011). From one vegetation type to the other, earthworm species composition may change (Margerie et al. 2001).
Another significant factor that affects the earthworm species composition is dis turbance, the major contributor being human activities (Callaham et al. 2003;Winsome et al. 2006). Land conversion or habitat transformation lead to decreased native assemblages and an increase in exotic species, since native species prefer less disturbed habitats (Win some et al. 2006).
Knowledge of the South African megadrile fauna is still incomplete. At present, 142 Microchaetidae species and 107 acanthodrilids (Acantho dri lidae: Acan tho drilinae) are known, all claimed as being indigenous to this country. Fifty species be longing to Acan thodrilidae (Acanthodrilinae, Benhamiinae), Eudrilidae (Eudri linae, Pareudrilinae), Glossoscolecidae, Lumbricidae, Megascolecidae and Oc nerodrilidae are known to be introduced (Plisko 2010). It is expected that yet more en demics and also alien species might be found in this part of the African continent. Re searchers have un dertaken work on South African earthworms since the late 1800s (Kinberg 1867;Beddard 1897;Michaelsen 1899), and there are numerous recent and current studies on different aspects of earthworms (e.g. Horn et al. 2007;Plisko 2010). However, the diversity, range and ecology of this special pedofaunal group still need attention.
In protected areas in South Africa, the size and composition of earthworm communities are not well known. Furthermore, the extent of invasions by introduced earthworms and their possible effects on indigenous earthworm species in protected areas have not been investigated. Research by Plisko (1995Plisko ( , 2000 shows the role of nature reserves in the protection of endemics, revealing that indigenous species are mostly limited to na tural habitats. Michaelsen (1913), Ljungström (1972a), Dlamini (2002) and Plisko (2010) reported on a large number of introduced species in South Africa, most of which originated from Europe, India or Asia. These alien species are common in all biotopes, agricultural land, grasslands, compost and forests, as well as in nature reserves.
The purpose of this paper is to extend study on earthworms in protected areas, observing co-existence between indigenous species and exotics. According to other surveys conducted in the Kruger National Park (Reinecke & Ackerman 1977;Plisko 1995) and Dlinza Forest (Plisko 2000), and occasional observations from Vernon Crookes Nature Reserve by Plisko (1992Plisko ( , 1998, endemics may survive by being protected from agricultural and industrial practices, even in presence of exotics.

Site description
The study was carried out in Queen Elizabeth Park, Pietermaritzburg, KwaZulu-Natal. It is a small (93 ha) park located on a north-facing hill of the Midlands mistbelt ex tension, in close proximity to the Pietermaritzburg city centre (ca 8 km north-west). The reserve was proclaimed in 1960 as a protected area; it had been an open park for public use (Anonymous 1989). The park contains research offices and laboratories, and conservation and administration departments of Ezemvelo KwaZulu-Natal Wildlife. This park is widely used as an outdoor education venue. The area has a sub-humid cli mate with wet summers and dry winters (average rainfall of 3.16 mm in January and 1.54 mm in February and average temperature of 15.7-27.64 °C; weather data are not avail able from the study area, so data from Cedara weather station were used in stead (Botes pers. comm. 2012)). Sampling was undertaken during the rainy season (January-February 2012). The elevation of the sampled area ranges from 2996 feet (913.2 m) to 3149 feet (959.8 m).
According to Le Roux (1988), the reserve consists of six main vegetation types, i.e. Grassland, Woodland, Lawn, Indigenous bush, Exotic Plantation and Lanes, and Mixed Scrub (Fig. 1). Being a protected area, this site was chosen to provide benchmark information because of minimal anthropogenic disturbance. It is possible that the park will retain its standard over time and remain suitable for further studies.

Material
Collections were made in all six vegetation types; five 1×1m quadrates were selected within each vegetation type, totalling 30 plots altogether. Earthworms were collected by digging and hand sorting. The average depth of digging was 0.5 m due to the presence of roots or rocks in the ground, and the soil was sorted for earthworms. All earthworms were narcotized using 45 % ethanol and then fixed in 4 % formaldehyde and preserved in 70 % ethanol. Specimens, except for juveniles, were examined and identified to species level using a Wild Heerbrugg microscope. For identifying Tritogenia, Plisko (1992) was used; and for Megascolecidae, Blakemore's (2004b) classification was followed. The material obtained was incorporated into the KwaZulu-Natal Museum collection. For comparison, type material borrowed from the Zoological Museum, Hamburg, was examined.

Synonymy/chresonymy and species descriptions
Introduced earthworm species are widely distributed, being known from all over the world, and they have numerous synonyms. In this paper, only distributional information and synonymy/chresonymy relevant to South Africa, are given. The descriptions of species are limited to the most important characters.

4
in the T. sulcata species-group, to which the following belong: T. sulcata Kinberg, 1867, T. howickiana (Michaelsen, 1913), T. debbieae Plisko, 2003, andT. hiltonia Plisko, 2003, all occurring in close proximity. The similar/distinguishing characters noted in these species are shown in Table 1. Because the type material of sulcata and howickiana ( Fig. 2) is in poor condition internally and comparative anatomical study may be not productive, the published data (Michaelsen 1899(Michaelsen , 1913Plisko & Zicsi 1991)  are considered. Material from QEP differs from type material of debbieae and hiltonia kept at the NMSA in thickness of the septa, shape and position of papillae, and number of spermathecae, indicating a close resemblance to the description of howickiana. Therefore, the specimens from QEP are assigned to this species. However, the known characters of the taxa in the sulcata-group are closely related (Table 1); a systematic revision based on molecular study is suggested.
Amynthas minimus (Horst, 1893)  Remarks: The large population of athecate specimens collected in QEP indicates parthe nogenetic reproduction. Ljungström (1972a) likewise collected parthenogenetic morphs with spermathecae and seminal vesicles reduced or absent, and also with an ab normally placed or shaped clitellum. However, thecate specimens have been recorded in this country as well (Plisko 2010;NMSAD). The species is known from various biotopes: grassland and meadows; and forested, natural and cultivated land (Zicsi 1998;Plisko 2010). It should be mentioned that thecate degeneration is a known phenomenon, attributed to parthenogenetic reproduction. Few species exhibiting it have been described (Gates 1932(Gates , 1972Tsai et al. 2002), although their systematic position has been discussed (Blakemore 2003;Gates 1972;Sims & Easton 1972). Some of the species have been im precisely accredited to a 'species group'.

DISCUSSION
The results from this survey add to understanding of the earthworm species composition in protected areas. Nine species were collected in Queen Elizabeth Park, one of which (Tritogenia howickiana) is indigenous, belonging to the family Microchaetidae.There is no history of its previous collection in the park, although in the close neighbourhood it was recorded in more than 20 sites. The 44 specimens (3.9 % of the total number collected) of T. howickiana were found in three of the six vegetation types (Table 2). These findings are in accordance with previous observations by other authors (i.e. Cal-laham et al. 2003;Winsome et al. 2006), who noted that indigenous earthworm species are limited in their distribution. The other 96.1 % of species belong to eight in troduced species of the families Acanthodrilidae (Benhamiinae), Glossoscolecidae, Me gascolecidae, and Lumbricidae. These eight species are well known among the world's widely distributed peregrine species (Hendrix & Bohlen 2002).
In QEP, seven species occur in large numbers in the indigenous bush and woodland; five species in mixed scrub, grassland and plantation each; and only four species in the lawn (Table 2). Species assemblages were dominated by Amynthas rodericensis, which appears to be by far the most common and widespread taxon, present in large num bers in all vegetation types, making up 34.4 % of the total number of earthworms (Table 2). Amynthas minimus was the next most common spe cies (18.0 % of the to tal number of earthworms), present in all vegetation types but in smaller numbers than A. ro dericensis. Pontoscolex corethrurus was found in three ve getation types, where as Di chogaster bolaui was seen in only one area in mixed scrub and was the least common. Introduced species dominated in the present study, com prising 96.1 % of found earthworms. The exotic species might be abundant due to their ability to re produce parthenogenetically. According to Ljungström (1972a), as well as Reynolds and Reinecke (1976), parthenogenetic species may have an advantage as re gards colonizing of new habitats.
The number of species collected in this study is similar to that previously recorded from other nature reserves (NMSAD); and according to Curry (1998), most earthworm communities are species-poor. Reynolds and Reinecke (1976) carried out an earthworm survey of the Kruger National Park and, surprisingly, found no native but only intro duced species. Zicsi and Reinecke (1992) recovered Nemertodrilus kruegeri and N. trans vaalensis but their endemism to South Africa is not clear (Plisko 2010). It is there fore of great interest that one indigenous species in QEP still survives, though in limited patches. The discovery of such an overwhelming abundance of introduced species in this protected area is not unanticipated after the findings of Reynolds and Reinecke (1976), particularly given the fact that Kruger National Park is a protected territory of national standard. During the course of extensive work on indigenous megadrile earthworms in South Africa, Plisko has noted a decline in the number of earthworm species due to change in the environment, be it development, agriculture or any other human activities that alter the natural habitats (Plisko pers. comm. 2012). Microchaetids occur mainly in grass lands and forests (Plisko 1997(Plisko , 2000(Plisko , 2003 and indigenous acanthodrilines are com monly restricted to soil that grows indigenous bushes (Pickford 1937;Ljungström 1972b;Plisko 2000). Introduced species have the ability to adapt in habitats that may be unfavourable, thus increasing their chances of being sampled.