Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tafz20

Download by: [Bibl de L'Univ de Bourgogne] Date: 02 November 2016, At: 00:49

African Zoology

ISSN: 1562-7020 (Print) 2224-073X (Online) Journal homepage: http://www.tandfonline.com/loi/tafz20

A new galago species for South Africa (Primates:Strepsirhini: Galagidae)

Fabien Génin, Ayabulela Yokwana, Nokuthula Kom, Sébastien Couette,Thibault Dieuleveut, Stephen D Nash & Judith C Masters

To cite this article: Fabien Génin, Ayabulela Yokwana, Nokuthula Kom, Sébastien Couette,Thibault Dieuleveut, Stephen D Nash & Judith C Masters (2016) A new galago species for SouthAfrica (Primates: Strepsirhini: Galagidae), African Zoology, 51:3, 135-143

To link to this article: http://dx.doi.org/10.1080/15627020.2016.1232602

Published online: 31 Oct 2016.

Submit your article to this journal

View related articles

View Crossmark data

African Zoology 2016, 51(3): 135–143Printed in South Africa — All rights reserved

Copyright © Zoological Society of Southern Africa

AFRICAN ZOOLOGYISSN 1562-7020 EISSN 2224-073Xhttp://dx.doi.org/10.1080/15627020.2016.1232602

African Zoology is co-published by NISC (Pty) Ltd and Taylor & Francis

Conservation of South Africa’s biodiversity requires the accurate identification of the units that comprise it. Most studies of biodiversity focus on species as the units of conservation, not only because they consist of lineages with independent evolutionary histories and fates (Simpson 1951; Wiley 1978), but also because they have unique suites of ecological adaptations and relationships to the environment (Schluter 2001). Species emergence and extinction are driven by environmental change, and speciation is an irreversible process that fixes the range of genetic variation and adaptive potential for a species’ metapopulation (Futuyma 1987). Hence, for conserva-tionists to construct an accurate picture of the environ-mental requirements for the survival of a community, it is crucial that they have the correct taxonomic information for that community.

For many species, morphology is a good indicator of specific identity, but this is not always the case. In nocturnal species such as galagos (bushbabies or nagapies), active when light intensity is low, the location, attraction and identification of potential mating partners is seldom dependent on visual signals, particularly those that are encoded into the animals’ morphology. Instead, communi-cation among species members is largely based on vocal

and olfactory signals. Speciation hence may not involve obvious changes in gross morphology, giving rise to cryptic species complexes (Henry 1985; Masters and Spencer 1989), or groups of reproductively independent species with very similar morphologies. Loud calls are by far the most accessible, reliable and non-invasive identifiers of galagid species (Masters 1993; Bearder et al. 1995; Masters and Couette 2015); indeed, galagos owe their colloquial name ‘bushbabies’ to the specific long-distance advertise-ment call of the thick-tailed greater galago, Otolemur crassicaudatus. Pelage colouration – one of the primary characters used by taxonomists to classify variation – is likely to have evolved in galagids under selective pressure for concealment from predators while the animals are resting during the daytime, rather than as an indicator of specific distinction. Hence, taxa such as Otolemur spp., which occupy less dense and highly seasonal forests that lose their leaves in the dry season, show discrete regional and probably seasonal patterns of colouration on the head, dorsum and tail, as well as in the colours of the hands and feet. Dwarf galagos, which inhabit coastal and/or evergreen forests, show much less regional variation, although there is a notable increase in rufous colouration the closer the animals are found to the Equator.

A new galago species for South Africa (Primates: Strepsirhini: Galagidae)

Fabien Génin1*, Ayabulela Yokwana1, Nokuthula Kom1, Sébastien Couette2, Thibault Dieuleveut3, Stephen D Nash4

and Judith C Masters1

1 African Primate Initiative for Ecology and Speciation, Department of Zoology and Entomology, University of Fort Hare, Alice, South Africa2 Ecole Pratique des Hautes Etudes, Laboratoire Paléobiodiversité et Evolution and UMR uB CNRS 6282 “Biogéosciences”, Université de Bourgogne, Dijon, France3 Emirates Center for Wildlife Propagation, Missour, Morocco4 10 Bayles Avenue, Stony Brook, New York, USA* Corresponding author, email: FGenin@ufh.ac.za

The primate fauna of South Africa has historically been viewed as comprising three diurnal cercopithecoid taxa – chacma baboons (Papio ursinus), vervet (Chlorocebus pygerythrus) and samango monkeys (Cercopithecus albogularis) – and two nocturnal lorisoid species – the thick-tailed greater galago (Otolemur crassicaudatus) and the southern lesser galago (Galago moholi). Here we report the positive identification of a third galago species within South Africa’s borders: the Mozambique dwarf galago or Grant’s galago, Galagoides granti (Thomas and Wroughton, 1907). The taxon was previously held to be restricted to Mozambique, eastern Zimbabwe, Malawi and Tanzania, but we have also observed it in the sand forest of Tembe Elephant Park and the Tshanini Community Reserve, near the Mozambique border. The species was formerly mistaken for Galago moholi, erroneously (we believe) extending the range of the latter species into northern KwaZulu-Natal. In South Africa the two small galagos are unlikely to have overlapping ranges: Galago moholi prefers dry savanna woodlands, whereas Galagoides granti is apparently confined to dry sand forest. However, both species may coexist with the larger and more widespread Otolemur crassicaudatus, an inhabitant of moist savanna, forest edge and thicket. The true South African ranges of both small galago species need to be ascertained.

Keywords: dwarf galagos, Galagoides, sand forest, Tembe, Tshanini

Introduction

Génin, Yokwana, Kom, Couette, Dieuleveut, Nash and Masters136

South Africa’s galagid fauna was previously held to comprise only the thick-tailed greater galago (Otolemur crassicaudatus) and the southern lesser galago (Galago moholi). The Mozambique dwarf galago or Grant’s galago (Galagoides granti; Figure 1) was previously believed to be restricted to Mozambique, the eastern highlands of Zimbabwe, Malawi and southern Tanzania (Smithers 1983; Skinner and Smithers 1990; Skinner and Chimimba 2005). Kyle (1996) reported sightings of Galago moholi in northern KwaZulu-Natal, a record that struck us as anomalous given the species’ preference for dry savanna woodland. More detailed surveys of Tembe Elephant Park and the nearby Tshanini Community Reserve revealed that the small-bodied galagids present in these areas are in fact Galagoides granti (Thomas and Wroughton, 1907). In this contribution we present the results of our first field study of this new addition to South Africa’s primate fauna. We abbreviate the genus Galagoides as Gs to distinguish it from the abbreviation used for Galago (G).

A brief systematic history of Grant’s galagoGalagoides granti was described as Galago granti by Thomas and Wroughton (1907: 286) as ‘a member of the moholi group, with an unusually bushy black-tipped tail.’ It was named for Captain CHB Grant who collected a series of eight specimens at Coguno, in Inhambane province in southern Mozambique, one of which became the type of the new taxon. In 1931, Ernst Schwarz, one of the twentieth century’s most fervent taxonomic lumpers, relegated Gs. granti to subspecies status within G. senegal-ensis (Schwarz 1931), a practice that was widely followed (Allen 1939; Roberts 1951; Hill 1953; Smithers and Tello 1976; Petter and Petter-Rousseaux 1979; Smithers 1983). Also subsumed under G. senegalensis by Schwarz (1931), Hill (1953) and subsequent authors were G. moholi and three taxa now considered to be allied with G. granti: Gs. cocos (from Kenya), Gs. nyasae (from Malawi) and Gs. zanzibaricus (from Zanzibar). Kingdon (1971) recognised Gs. zanzibaricus as a distinct species, a proposal supported by Groves (1974) who suggested that Gs. granti was likely to be the southern representative of the Zanzibar species. Galagoides zanzibaricus granti was the taxonomy followed for more than 20 years (Meester et al. 1986; Jenkins 1987; Skinner and Smithers 1990; Groves 1993). Following recommendations by Bearder et al. (1996), Honess and Bearder (1996) and Kingdon (1997), Gs. granti has been treated more recently as a distinct species within the Zanzibar group (Masters and Bragg 2000; Groves 2001; Grubb et al. 2003; Groves 2005; Butynski et al. 2006, 2013; Nekaris 2013; Masters and Couette 2015; Pozzi et al. 2015).

Methods

Study sitesCaptain Grant recorded the following observation of the Gs. granti type locality: ‘Inland from Inhambane the country is composed of more or less undulating, sandy flats, densely bushed and timbered’ (Thomas and Wroughton 1907: 285). This description fits Tembe Elephant Park and Tshanini Community Reserve equally well, and inspired

our initial search for dwarf galagos in the reserves. The region consists of flood plains based on clay, covered in parts with ancient, parallel, east–west-orientated sand dunes – the relicts of an ancient sea shore. It forms the southernmost extent of a much larger domain extending well into Mozambique. The vegetation is a complex mosaic of dry forest dominated by large endemic sand forest newtonias (Newtonia hildebrandtii), thicket, woodland

Figure 1: Adult Galagoides granti in Tshanini Community Reserve. Photo by Hajarimanitra Rambeloarivony

African Zoology 2016, 51(3): 135–143 137

and savanna grassland. Similar, naturally fragmented sand forests occur in a few other localities in northern KwaZulu-Natal, sometimes in protected areas (e.g. Mkuze and Ndumo). They form the southern limit of distribu-tion for a number of locally rare species, such as the suni (Neotragus moschatus), the four-toed elephant shrew (Petrodromus tetradactylus) and the plain-backed sunbird (Anthreptes reichenowi).

Preliminary surveyIn this contribution we report the results of preliminary observations carried out from 2011 to 2014. We observed Grant’s galagos only in the dry sand forest and dense woodland of Tembe Elephant Park (six nights) and the Tshanini Community Reserve (12 nights). We conducted our initial studies in Tembe Elephant Park (27.0486° S, 32.4222° E). While populations of both Otolemur crassi-caudatus and Gs. granti were both accessible and relatively abundant, so were the populations of lions (Panthera leo) and elephants (Loxodonta africana), which hindered our nocturnal surveys. We therefore moved our field site to Tshanini Community Reserve, which is situated approxi-mately 6 km to the south of the southern border of Tembe Elephant Park (Figure 2; see also Gaugris et al. 2004). Our observations focused on sleeping sites, because they form the focus of activity at the beginning and end of the galagos’ nocturnal activity period.

Recording methodsAnimals were first recorded opportunistically, between January 2012 and March 2014, and then systemati-cally, during the population density survey in September 2014. We recorded vocalisations using a SEINHEISER MKE600 semi-directional microphone and a ROLAND R-26 digital recorder. We focused on three selected calls, most frequently emitted by Gs. granti and well described in the literature: the incremental loud call regarded as specifically diagnostic (Bearder et al. 1995), and the more conserved ‘buzz screech’ and ‘yap’, used as alarm calls by other Galagoides and Galago species.

Population surveyBecause of the complexity of the habitat, we surveyed portions of a single 9.6 km transect twice daily (evening and morning) that crossed three subhabitats: (1) sand forest: two patches of tall sand forest and a portion of low sand forest and thicket; (2) savanna: Terminalia–Combretum savanna grassland; and (3) woodland: a dense woodland rich in creepers and lianas on clay marking a transition between the first two subhabitats along the dry bed of an ancient river. The 9.6 km transect was divided into three 3.2 km parts that could be covered in a single survey period. To avoid double counts, the boundaries between survey sectors were located in areas of unsuitable habitat (open grassland), and each portion had a different subhabitat constitution (dense sand forest, dense woodland, mosaic of sand forest and grassland–savanna).

To optimise the chance of recording the animals, we conducted recording surveys over a period of 6 d focusing on the times when galagos are known to be particularly active and vocal (Harcourt and Nash 1986), i.e. the first and

the last hour of the nocturnal activity period. Over these 6 d, we conducted a total of 12 h of early evening surveys (between 18:00 and 22:00), 12 h of early morning surveys (between 02:00 and 06:00), and a single 6 h diurnal survey (06:00–12:00) for a comparison of background noise and to mark the survey path with biodegradable tape. We walked the path at an average speed of approximately 1 km h−1, stopping every 100 m to record for 5 min. The time taken to cover each 3.2 km survey sector was approxi-mately 4 h. We also recorded galagos we encountered as we moved, identified either by calls or by the reflection of the tapetum in the light of our headlamps. The recording equipment was turned on each time an animal was located, and we approached the subjects when possible. We recorded the GPS coordinates and height above ground of the animals when they were first sighted. In addition, we measured the distance from the observer to the occupied tree using a BOSCH 80 laser telemeter, to assess visibility and correct our estimates of population density in the

AFRICA

SouthAfrica

SOUTHAFRICA

Conservation/Protected area

Road (N2)

0 50 100 km

ThembeElephant

Park

TshniniGame Reserve

Mozambique

Swaziland

Pongola

Mkuze

Ulundi

Richards Bay

Durban

INDIAN OCEAN

KWAZULU-NATAL

32° E 33° E

29° S

28° S

27° S

Figure 2: Map showing the sites where Galagoides granti has been observed in South Africa. Tembe Elephant Park and Tshanini Community Reserve are indicated by arrows

Génin, Yokwana, Kom, Couette, Dieuleveut, Nash and Masters138

heterogeneous habitat: the maximum distance of visual detection measured in each subhabitat was used to calculate the width of the survey path and convert linear abundance in individuals km–1 into surface-based popula-tion density.

Results

Sleeping sites and general observations In 24 nights of surveys we identified six sleeping sites, four of which were on the population density survey transect. All were located in deep tree holes in tall sand forest habitat, situated 3–14 m above ground in living (n = 5) or dead (n = 1) trees. Of these six sites, four were occupied by groups of three females, one by a female and an infant (Figure 3), and one by a male. We also found an unoccupied galago branch-nest (probably built by greater galagos) in woodland habitat, and three O. crassicaudatus sleeping sites, all in large savanna trees.

We observed five social encounters between female dwarf galagos during the night, which were accompanied by social calls (descending screech, chatter). Foraging animals, however, were generally observed alone (n = 9). When foraging, dwarf galagos used all vegetation strata, from the upper canopy to the leaf litter, where they were seen foraging for insects (n = 4). They were also observed feeding on Combretum molle and Sclerocarya birrea gum (n = 3) and on the nectar of the introduced Agave sisalis (n = 1).

Population densitiesThe results of our transect surveys are presented in Table 1. Our values for Gs. granti are similar to other abundance estimates for the species in East Africa (Honess et al. 2013), although linear rates of encounter are not directly comparable because the method is very sensitive to variation in visibility. Our data indicate that the overall population density of the thick-tailed greater galago is much higher than that of the Mozambique dwarf galago, but this is essentially due to the higher density of its preferred habitat in the subhabitat mosaic. Indeed, Gs. granti was more abundant than O. crassicaudatus within its own preferred habitat of tall sand forest. This habitat preference is likely to be related to the species’ use of large, hollow trees as sleeping sites. Dwarf galagos do venture into more open habitat (seven observations), but only later in the night when foraging. A similar use of habitat is seen in the closely related Gs. zanzibaricus (FG pers. obs.).

VocalisationsDuring the population survey, animals were more often detected by their calls (n = 25) than by their eye shine (n = 6). Figure 3 presents comparative sonagrams of Gs. granti calls recorded at Tshanini Community Reserve with various other galago calls, including Gs. granti calls recorded in southern Tanzania and vocalisations emitted by G. moholi. Three calls were selected for analysis on the basis that they were most frequently emitted.

Loud/advertisement callsGalago loud calls are known to be highly reliable and readily accessible indicators of species identity (Zimmermann 1990;

Masters 1993; Bearder et al. 1995, 1996). The Mozambique dwarf galago has a high-frequency, modulated loud call typical of the Galagoides species east of the African Rift. This type of call contrasts strongly with the low-frequency, almost tonal loud call of G. moholi. Mozambique dwarf galagos belong to the cluster of so-called ‘incremental callers’ (Bearder et al. 1995; Kingdon 1997): the loud calls are made up of repeated short units given in groups of increasing number (increments) (see Figure 3). By contrast, southern lesser galagos utter a metronomic two-unit bark with a shorter second unit (‘repetitive callers’).

Incremental calls are most often used as cohesive gathering calls when animals exit (n = 12) or return to (n = 7) their sleeping sites. They are also emitted sporadically throughout the night, probably when animals meet the scent marks of nearby or affiliated individuals (n = 3). Such calls always elicited responses from nearby individuals.

Buzz screech alarmThe one-unit buzz screech used by the Mozambique dwarf galago is similar and probably homologous to the chow-ha alarm call given by G. moholi. The call is shared by all members of the Gs. zanzibaricus species complex, and is reminiscent of the two-unit, more explosive alarm call emitted both by Gs. demidoff and Gs. orinus. The buzz screech is rather stereotyped but of variable duration, sometimes uttered in isolation (in response to a low level of disturbance), but more often associated with yaps, and sometimes with more complex calls (chatter or irregular emission of short, modulated units and double units resembling the basic units of the loud call).

Mobbing yapsIn contrast, the mobbing yap is the most phylogenetically conserved of all galago vocalisations. It is not peculiar to the Galagoides group, and cannot be used to distinguish the two small galagos found within South Africa (the apparent differences observed in Figure 3 are probably an artefact of sound degradation affecting the transmission of higher frequencies, as the maximum energy occurs around 4 kHz in all species). The yap serves as a mobbing call and is sometimes used in conjunction with the less frequently emitted buzz screech (n = 5). It was emitted in the presence of a ‘parked’ infant, or when an animal’s path was blocked by a human observer.

Discussion

The recent discovery of Grant’s galago within the borders of South Africa carries the implication that previous observations of small-bodied galagos in northern KwaZulu-Natal may have been mistakenly attributed to G. moholi, erroneously extending the distribution of this species into the extreme north-eastern part of South Africa. We hence have a situation in which one of South Africa’s relatively common small mammal species, the lesser galago or nagapie, has suddenly become data deficient, as we do not know the true extent of its distribution. We are similarly ignorant of the extent of the distribution of Gs. granti within the country’s borders. A pressing research need is thus to determine the ranges of these two species, a task which

African Zoology 2016, 51(3): 135–143 139

Gal

agoi

des

gran

ti (T

shan

ini):

buzz

scr

eech

Gal

agoi

des

gran

ti (T

shan

ini):

yaps

Gal

agoi

des

gran

ti (T

anza

nia)

:in

crem

enta

lG

alag

oide

s za

nzib

aric

us (T

anza

nia)

:bu

zz s

cree

ch a

nd y

aps

Gal

agoi

des

rond

oens

is (T

anza

nia)

:ya

ps

Gal

ago

moh

oli (

Zam

bia)

:re

petit

ive

bark

Gal

agoi

des

dem

idof

f (C

amer

oon)

:tw

o-un

it bu

zz s

cree

ches

and

yap

sG

alag

o m

ohol

i (Za

mbi

a):

yaps

LOU

D C

ALLS

ALAR

M C

ALLS

MO

BBIN

G C

ALLS

Gal

agoi

des

gran

ti (T

shan

ini):

incr

emen

tal

Figu

re 3

: Com

para

tive

sono

gram

s of

eas

tern

and

wes

tern

Gal

agoi

des

spp.

and

less

er g

alag

os (G

alag

o m

ohol

i), s

how

ing

the

thre

e ca

ll ty

pes

utte

red

mos

t fre

quen

tly b

y G

alag

oide

s gr

anti:

th

e in

crem

enta

l lou

d ca

ll, th

e bu

zz s

cree

ch a

nd th

e m

obbi

ng y

ap

Génin, Yokwana, Kom, Couette, Dieuleveut, Nash and Masters140

requires input from field researchers, wildlife conserva-tionists and enthusiasts, in addition to our own continued surveys. To assist field identification of the two small galago species, we provide a summary of morphological and behavioural clues to their respective identities.

Distinguishing South Africa’s two smaller galago speciesThe two species are very similar in body size; compara-tive measurements gathered from specimens stored in museum collections worldwide are presented in Table 2. The major features distinguishing Gs. granti from G. moholi are illustrated in Figure 4. Galagoides granti is generally darker and browner than G. moholi with a dark bushy tail (see also Figure 1), whereas the dominant colour on the head, dorsum and outer surfaces of the limbs of G. moholi is pale grey washed with varying degrees of russet brown, and the tail is slender and coated with a grizzled mix of brown and grey hairs; the limbs are washed with pale yellow. The ventral surface of Gs. granti is yellow to cream-buff, in contradistinction to the white ventrum of G. moholi. The shape of the snout is diagnostic. In Grant’s galago it is narrow and pointed, and the nasal profile is concave (Figure 5), whereas lesser galagos have a short, square muzzle and a straight nasal profile. In both taxa the eyes are surrounded by black circumocular rings, which contrast with a white to pale grey stripe that traverses the length of the nose, but in lesser galagos the face is more or less suffused with white.

The skulls of the two taxa are readily distinguished. Galagoides spp. have a relatively large, unreduced M3 that approximates the molarised P4 in size, whereas the same tooth is much reduced in Galago spp., in correspondence with their shorter snouts. The difference in the nasal profiles is very evident (see illustrations in Smithers 1983; Skinner and Smithers 1990; Skinner and Chimimba 2005). In Galagoides spp. the premaxillary bones are extended to form a tube that causes the upper jaw to project beyond the lower jaw; in Galago spp. the premaxillaries are not extended beyond a relictual nub on the midline beneath the

nasal opening, suggesting that the extended condition is ancestral. Galagoides spp., as befits their locomotor mode, have relatively unflexed basicrania and flatter braincases, whereas the skulls of Galago taxa are more globular, with a higher braincase and a more flexed basicranium.

In the field Galagoides spp. are more quadrupedal than Galago spp. When they leap from tree to tree, they make first contact with the landing surface either with the hands or with all four extremities together. Galago spp., by contrast, are specialised ricochetal leapers (Oxnard et al. 1990). They swing their hind limbs forward during the course of the leap and land feet first. This more derived locomotion, associated with much longer hind limbs and a more globular skull, is an adaptation to moving through relatively open savanna woodland habitat, where the trees are widely spaced.

The most reliable character for field identification of Grant’s galago is its distinctive vocal repertoire. The calls given most frequently, particularly in the presence of an observer, are the buzz screech and clearer, high-pitched yaps. Similar calls are emitted by G. moholi, however, and they are not informative of species identity to unfamiliar ears. The loud advertisement call, in contrast, distinguishes the two small galagos in South Africa unambiguously. The loud call of Gs. granti is described as incremental because new modulated units are added in a complex sequence. This call, uttered more frequently near sleeping sites and sporadically throughout the night, contrasts markedly with the regularly repeated, monotonous, lower-frequency tonal loud call signalling the presence of G. moholi.

Finally, the two small galago species also differ signifi-cantly in their preferred habitats, which are unlikely to coexist within the boundaries of South Africa. The southern lesser galago is an inhabitant of dry savanna woodland, often riverine woodlands in the driest parts of their range, whereas the Mozambique dwarf galago is a dry forest and dense woodland specialist. Both species may coexist locally with the much larger thick-tailed greater bushbaby (Otolemur crassicaudatus), which is found in a variety of

SpeciesLinear density (encounters km−1) Population density (strip width method) (individuals km−2)

Sand forest Woodland Savanna Sand forest Woodland SavannaGalagoides granti 1.3 1.4 0.2 90.9 21.3 4.2Otolemur crassicaudatus 0.8 2.8 2.1 54.5 42.7 40.7

Table 1: Comparison of Galagoides granti and Otolemur crassicaudatus abundance in Tshanini Community Reserve in three kinds of habitat

Average measurement Galagoides granti Galago moholiHead-body length (mm) 153.7 ± 1.64 (n = 50) 147.7 ± 1.46 (n = 50)Tail length (mm) 220.5 ± 2.39 (n = 50) 223.7 ± 1.73 (n = 50)Hind foot length (mm) 58.5 ± 0.54 (n = 50) 57.2 ± 0.68 (n = 50)Ear length (mm) 38.2 ± 0.41 (n = 50) 36.6 ± 0.43 (n = 50)Body weight (g) 146.8 ± 7.14 (n = 20) 153.5 ± 2.52 (n = 50)Total skull length (mm)

(prosthion to opistocranion) 42.3 ± 0.24 (n = 38) 39.4 ± 0.19 (n = 50)

Table 2: Average body and skull measurements (±SE) of museum specimens of Galagoides granti and Galago moholi. General body measurements were taken by the collectors on freshly killed specimens and recorded by us from museum records. Skull lengths were measured by us using digital callipers to 0.1 mm accuracy

African Zoology 2016, 51(3): 135–143 141

habitats (moist savanna woodland, thicket and forest edge).

Habitat preferences and biogeography of South African galagosLike other eastern dwarf galagos, Gs. granti prefers well-wooded habitat for sleeping sites, although the animals venture into more open habitat while foraging. Galagoides

granti has a clear predilection for mosaics of thicket and tall, mature woodland rich in creepers and lianas, connected to patches of tall dry forest used exclusively during the daytime resting period. The range of the South African population is limited in the west by the eastern escarpment and in the east by the Indian Ocean coastline, confirming a preference for subtropical conditions.

Linder et al. (2012) compared the distributions of plants and vertebrates in sub-Saharan Africa, and identified five or six distinct biogeographic regions on the subcontinent, depending on the taxa examined. The ranges of the three South African galago species match two of these regions, but no two species have the same habitat preference, implying that the three lineages colonised South Africa via distinct routes and probably at different times (Pozzi 2016). Linder and colleagues identified a distinct Mozambique biogeographic domain characterised by many endemic plants and birds, but few endemic mammals. In South Africa the Mozambique region is probably confined to the Tembe–Tshanini area, the vast majority of it occurring further north.(1) Galagoides granti has an exclusively Mozambique

distribution that reaches the western foot of the eastern escarpment in Zimbabwe and Malawi. It is a rare forest and dense woodland specialist localised in South Africa, where we have observed it only in the Tembe and Tshanini reserves, which are hence likely to mark the southern extreme of the Gs. granti distribution (Figure 2).

(2) Galago moholi has a restricted dry Zambezian

Figure 4: Morphological comparison of Galagoides granti and Galago moholi. Drawings by Stephen D Nash

Figure 5: Lateral view of a Galagoides granti juvenile showing the concave nasal profile, the extended upper jaw, and the contrasting eye rings and nasal stripe. The eye rings darken in maturity

Génin, Yokwana, Kom, Couette, Dieuleveut, Nash and Masters142

distribution excluding the Mozambique domain and matching the distribution of much more open savanna woodland and miombo. In South Africa, it is restricted to flood plains and riverine Vachellia karroo (Acacia karoo) woodland, from Pretoria to Limpopo province.

(3) Otolemur crassicaudatus has a broad Zambezian distri-bution including the Mozambique domain. However, its distribution is patchy and follows deep valleys where its preferred habitat of moist savanna woodland, forest edge and thicket occurs. In South Africa, the species is restricted to areas of Limpopo and Mpumalanga provinces and subtropical KwaZulu-Natal. It coexists locally with G. moholi and Gs. granti in the north and the north-east of its distribution, respectively.

The KwaZulu-Natal coastal region is characterised by subtropical transitional conditions, indicative of climatic fluctuations in the recent past that have left behind a great number of isolated, relictual populations of tropical species (e.g. the green barbet, Stactolaema olivacea). The Mozambique dwarf galago is an addition to this already impressive list. The northern sand forest has many other rarities of Zambezian (suni, four-toed elephant shrew, red squirrel and African broadbill), or Mozambican (plain-backed sunbird) affinities. Further investigations will include a thorough survey of all the sizeable sand forest fragments in northern KwaZulu-Natal, with a special emphasis on other types of sand forest (e.g. the Mkuze and Ndumo riverine sand forest and gallery forest). Such assessments would be particularly valuable as the presence of a rare, CITES II protected species may help to enhance the conservation status of the region. The extreme north of Maputuland is poorly developed and deserves more attention, especially targeting the conserva-tion of forest habitat.

Acknowledgements — We acknowledge the important contribu-tions made by Prof. Dai Harbert, who willingly shinned up trees and faced down elephants in the interests of science, and Greg Davies, whose pursuit of sunbirds was crucial to the discovery of dwarf galagos in South Africa. Dr Iris Dröscher, Hajarimanitra Rambeloarivony and Tarik Bodasing showed similar fortitude, and we are grateful to them for sharing their expertise and their photographs. The project was funded by National Research Foundation grants 93924 awarded to FG, and 92541 and 90772 awarded to JCM; grant number GB-TAP 4120 awarded by the Synthesys Program to SC; and an ABIC grant awarded to JCM by the Royal Museum for Central Africa through a Framework Agreement with the Belgian Development Co-operation.

References

Allen GM. 1939. A checklist of African mammals. Bulletin of the Museum of Comparative Zoology 83: 1–763.

Bearder SK, Honess PE, Ambrose L. 1995. Species diversity among galagos with special reference to mate recognition. In: Alterman L, Doyle GA, Izard MK (eds), Creatures of the dark: the nocturnal prosimians. New York: Plenum Press. pp 331–352.

Bearder SK, Honess PE, Bayes M, Ambrose L, Anderson M. 1996. Assessing galago diversity – a call for help. African Primates 2: 11–15.

Butynski TM, de Jong YA, Perkin AW, Bearder SK, Honess PE. 2006. Taxonomy, distribution, and conservation status of three species of dwarf galagos (Galagoides) in eastern Africa. Primate

Conservation 21: 63–79.Butynski T, Kingdon J, Kalina J. 2013. Mammals of Africa, vol. II:

Primates. London: Bloomsbury.Futuyma DJ. 1987. On the role of species in anagenesis. American

Naturalist 130: 465–473.Gaugris JY, Matthews WS, van Rooyen MW, Botma J du P. 2004.

The vegetation of Tshanini Game Reserve and a comparison with equivalent units in the Tembe Elephant Park in Maputaland, South Africa. Koedoe 47: 9–29.

Groves CP. 1974. Taxonomy and phylogeny of prosimians. In: Martin RD, Doyle GA, Walker AC (eds), Prosimian biology. London: Duckworth. pp 449–473.

Groves CP. 1993. Order Primates. In: Wilson DE, Reeder DM (eds), Mammal species of the world: a taxonomic and geographic reference (2nd edn). Washington, DC: Smithsonian Institution Press. pp 243–277.

Groves CP. 2001. Primate taxonomy. Washington, DC: Smithsonian Institution Press.

Groves CP. 2005. Order Primates. In: Wilson DE, Reeder DM (eds), Mammal species of the world: a taxonomic and geographic reference (3rd edn). Baltimore: Johns Hopkins University Press. pp 111–184.

Grubb P, Butynski TM, Oates JF, Bearder SK, Disotell TR, Groves CP, Struhsaker TT. 2003. Assessment of the diversity of African primates. International Journal of Primatology 24: 1301–1357.

Harcourt CS, Nash LT. 1986. Social organization of galagos in Kenyan coastal forests. I. Galago zanzibaricus. American Journal of Primatology 10: 339–355.

Henry CS. 1985. Sibling species, call differences, and speciation in green lacewings (Neuropoda: Chrysopidae: Crysoperla). Evolution 39: 965–984.

Hill WCO. 1953. Primates: comparative anatomy and taxonomy, vol. I: Strepsirhini. Edinburgh: University of Edinburgh Press.

Honess PE, Bearder SK. 1996. Descriptions of the dwarf galago species of Tanzania. African Primates 2: 75–79.

Honess PE, Bearder SK, Butinsky TM. 2013. Galagoides granti Mozambique dwarf galago (Grant’s dwarf galago). In: Butynski T, Kingdon J, Kalina J (eds), Mammals of Africa, vol. II: Primates. London: Bloomsbury. pp 454–456.

Jenkins P. 1987. Catalogue of primates in the British Museum (Natural History). London: British Museum (Natural History).

Kingdon J. 1971. East African mammals: an atlas of evolution in Africa, vol. I: Primates. London: Academic Press.

Kingdon J. 1997. The Kingdon field guide to African mammals. London: Academic Press.

Kyle R. 1996. Sight records of Galago moholi, the lesser bushbaby, in KwaZulu-Natal. Lammergeyer 44: 39–40.

Linder HP, de Klerk HM, Born J, Burgess ND, Fjeldsa J, Rahbek C. 2012. The partitioning of Africa: statistically defined biogeographical regions in sub-Saharan Africa. Journal of Biogeography 39: 1189–1205.

Masters JC. 1993. Primates and paradigms: problems with the identification of genetic species. In: Kimbel WH, Martin LB (eds), Species, species concepts, and primate evolution. New York: Plenum Press. pp 43–64.

Masters JC, Bragg NP. 2000. Morphological correlates of speciation in bush babies. International Journal of Primatology 21: 793–813.

Masters JC, Couette S. 2015. Characterizing cryptic species: a morphometric analysis of craniodental characters in the dwarf galago genus Galagoides. American Journal of Physical Anthropology 158: 288–299.

Masters JC, Spencer HG. 1989. Why we need a new genetic species concept. Systematic Zoology 38: 270–279.

Meester J, Rautenbach IL, Dippenaar NJ, Baker CM. 1986. Classification of southern African mammals. Transvaal Museum Monographs 5: 1–359.

African Zoology 2016, 51(3): 135–143 143

Nekaris KAI. 2013. Family Galagidae (galagos). In: Mittermeier RA, Rylands AB, Wilson DE (eds), Handbook of the mammals of the world, part 3: Primates. Barcelona: Lynx Edicions. pp 184–209.

Oxnard CE, Crompton RH, Lieberman SS. 1990. Animal lifestyles and anatomies: the case of prosimian primates. Seattle: University of Washington Press.

Petter J-J, Petter-Rousseaux A. 1979. Classification of the prosimians. In: Doyle GA, Martin RD (eds), The study of prosimian behavior. New York: Academic Press.

Pozzi L. 2016. The role of forest expansion and contraction in species diversification among galagos (Primates: Galagidae). Journal of Biogeography 43: 1930–1941.

Pozzi L, Nekaris KAI, Perkin A, Bearder SK, Pimley ER, Schulze H, Streicher U, Nadler T, Kitchener A, Zischler H, Zinner D, Roos C. 2015. Remarkable ancient divergences amongst neglected lorisiform primates. Zoological Journal of the Linnean Society 175: 661–674.

Roberts A. 1951. The mammals of South Africa. Pretoria: Trustees of “The Mammals of South Africa" Book Fund.

Schluter D. 2001. Ecology and the origin of species. Trends in Ecology and Evoloution 16: 372–380.

Schwarz E. 1931. On the African long-tailed lemurs or galagos. Annals and Magazine of Natural History 10(7): 41–66.

Simpson GG. 1951. The species concept. Evolution 5: 285–298.Skinner JD, Chimimba CT. 2005. The mammals of the southern

African subregion (3rd edn). Cambridge: Cambridge University Press.

Skinner JD, Smithers RHN. 1990. The mammals of the southern African subregion (2nd edn). Pretoria: University of Pretoria Press.

Smithers RHN. 1983. The mammals of the southern African subregion. Pretoria: University of Pretoria Press.

Smithers RHN, Tello JLP. 1976. Check list and atlas of the mammals of Mozambique. National Museums and Monuments of Rhodesia, Museum Memoir 8. Salisbury: Trustees of the National Museums and Monuments of Rhodesia.

Thomas O, Wroughton RC. 1907. The Rudd Exploration of South Africa. VII. List of mammals obtained by Mr. Grant at Coguno, Inhambane. Proceedings of the Zoological Society of London 1907: 285–299.

Wiley EO. 1978. The evolutionary species concept reconsidered. Systematic Zoology 27: 17–26.

Zimmermann E. 1990. Differentiation of vocalizations in bush- babies (Galaginae, Prosimiae, Primates) and the significance for assessing phylogenetic relationships. Zeitschrift für zoologische Systematik und Evolutionsforschung 28: 217–239.

Received 8 January 2016, revised 29 August 2016, accepted 1 September 2016Associate Editor: Ara Monadjem