MT MABU, MOZAMBIQUE: - Royal Botanic Gardens, Kew

Darwin Initiative Award 15/036: Monitoring and Managing 

Biodiversity Loss in South-East Africa's Montane Ecosystems 

MT MABU, MOZAMBIQUE: 

BIODIVERSITY AND CONSERVATION 

November 2012 

Jonathan Timberlake, Julian Bayliss, Françoise Dowsett-Lemaire, 

Colin Congdon, Bill Branch, Steve Collins, Michael Curran, Robert J. Dowsett, 

Lincoln Fishpool, Jorge Francisco, Tim Harris, Mirjam Kopp & Camila de Sousa 

ABRI 

african butterfly research in 

Forestry Research 

Institute of Malawi

Biodiversity of Mt Mabu, Mozambique, page 2 

Front cover: Main camp in lower forest area on Mt Mabu (JB). 

Frontispiece: View over Mabu forest to north (TT, top); Hermenegildo Matimele plant collecting (TT, 

middle L); view of Mt Mabu from abandoned tea estate (JT, middle R); butterflies (Lachnoptera ayresii) 

mating (JB, bottom L); Atheris mabuensis (JB, bottom R). 

Photo credits: JB – Julian Bayliss CS ‒ Camila de Sousa 

JT – Jonathan Timberlake TT – Tom Timberlake 

TH – Tim Harris 

Suggested citation: Timberlake, J.R., Bayliss, J., Dowsett-Lemaire, F., 

Congdon, C., Branch, W.R., Collins, S., Curran, M., Dowsett, R.J., Fishpool, 

L., Francisco, J., Harris, T., Kopp, M. & de Sousa, C. (2012). Mt Mabu, 

Mozambique: Biodiversity and Conservation. Report produced under the 

Darwin Initiative Award 15/036. Royal Botanic Gardens, Kew, London. 94 pp.


LIST OF CONTENTS 

List of Contents .......................................................................................................................... 3 

List of Tables .............................................................................................................................. 4 

List of Figures ............................................................................................................................ 5 

SUMMARY ............................................................................................................................... 6 

1. INTRODUCTION ................................................................................................................. 8 

1.1 Background .................................................................................................................... 8 

1.2 Media Coverage.............................................................................................................. 9 

2. DESCRIPTION OF STUDY AREA ................................................................................... 10 

2.1 Geography and Geology ............................................................................................... 10 

2.2 Climate ......................................................................................................................... 11 

2.3 Aerial Photos ................................................................................................................ 13 

3. HISTORY AND EARLY COLLECTING .......................................................................... 13 

3.1 Early History ................................................................................................................ 14 

3.2 Recent History .............................................................................................................. 16 

4. VEGETATION .. ................................................................................................................. 19 

4.1 Previous Studies ........................................................................................................... 19 

4.2 Vegetation Mapping ..................................................................................................... 19 

4.3 Vegetation Types .......................................................................................................... 22 

4.3.1 Plantations ............................................................................................................. 22 

4.3.2 Woodland .............................................................................................................. 22 

4.3.3 Moist Forest .......................................................................................................... 23 

4.3.4 Montane shrubland ............................................................................................... 27 

4.3.5 Vegetation of the Drier Western and Northern Slopes ......................................... 29 

4.4 Forest Plots ................................................................................................................... 29 

4.5 Observations on Tree Fruiting ...................................................................................... 29 

5. . PLANTS .............................................................................................................................. 33 


5.2 Plant Collections ........................................................................................................... 33 

5.3 Species of Particular Interest ........................................................................................ 34 

6. BIRDS ................................................................................................................................. 37 


6.2 Bird Survey ................................................................................................................... 37 

6.3 Biogeographical Significance ....................................................................................... 37 

6.4 Bird Species of Conservation Importance .................................................................... 39 

6.5 Bird Breeding Records ................................................................................................. 41 

7. OTHER VERTEBRATES & INVERTEBRATES ............................................................. 43 


7.2 Larger Mammals .......................................................................................................... 43 

7.3 Bats ............................................................................................................................... 43 

7.4 Other Small Mammals .................................................................................................. 44 

7.5 Herpetofauna ................................................................................................................ 45 

7.6 Lepidoptera ................................................................................................................... 47 

7.7 Molluscs ....................................................................................................................... 49 

7.8 Freshwater Crustacea.................................................................................................... 49 

7.9 Biogeography ............................................................................................................... 50


8. CONSERVATION .............................................................................................................. 51 

8.1 Conservation Threats .................................................................................................... 51 

8.2 Protection and Conservation......................................................................................... 54 

9. MAIN FINDINGS ............................................................................................................... 56 

10. RECOMMENDATIONS ................................................................................................... 57 

11. ACKNOWLEDGEMENTS ............................................................................................... 59 

12. BIBLIOGRAPHY & REFERENCES ............................................................................... 60 

Annex 1. Participants on main Mt Mabu, October 2008 ......................................................... 66 

Annex 2. Media Coverage of Mt Mabu ................................................................................... 67 

Annex 3. Plant checklist for Mt Mabu above 800 m ............................................................... 71 

Annex 4. Annotated list of bird species on Mt Mabu recorded above 400 m .......................... 77 

Annex 5. Birds ringed near main forest camp on Mt Mabu, 16‒17 Oct 2008 ......................... 84 

Annex 6. List of larger mammal species recorded from the Mabu massif .............................. 85 

Annex 7. List of small mammal species collected or recorded from the Mabu massif ........... 87 

Annex 8. List of reptile and amphibian species collected or recorded from the Mabu massif 88 

Annex 9. Butterfly species collected on Mt Mabu from 2006 to 2010 .................................... 89 

LIST OF TABLES 

Table 1. Main visits to Mt Mabu, 2005‒2010 ............................................................................ 9 

Table 2. Extent of area above 1000 m by altitudinal class for the Mt Mabu massif ................ 10 

Table 3. Tea enterprises in the Mt Mabu area, with 1960 production figures ......................... 16 

Table 4. Extent of forest area by altitudinal class for the Mt Mabu massif ............................. 20 

Table 5. Summary of forest plot data, Mt Mabu ...................................................................... 31 

Table 6. Plant species of interest recorded from Mt Mabu ...................................................... 34 

Table 7. Species found on Mt Mabu mentioned on Sabonet Red Data lists as threatened 

or endemic .................................................................................................................. 36 

Table 8. Afromontane (near-)endemic and selected Eastern endemic bird species present 

on Namuli, Mulanje, Thyolo, Mabu and Chiperone Mountains ................................ 38 

Table 9. Location of bat sampling sites and sampling effort ................................................... 44 

Table 10. List of molluscs collected on Mt Mabu, 2005‒2009 ................................................ 49


LIST OF FIGURES 

Fig. 1. Montane areas in N Mozambique studied under the Darwin Initiative project ............ 11 

Fig. 2. Google Earth image of Tacuane area showing Mt Mabuand roads (Aug 2006) .......... 11 

Fig. 3. Map of Mt Mabu area showing forest extent and main localities visited ..................... 12 

Fig. 4. Google Earth image of Mt Mabu massif from 2006 showing forest + dense woodland 

cover and abandoned tea plantations; and false-colour Landsat image from 2005 with 

dense vegetation cover ................................................................................................. 13 

Fig. 5. Map showing route of Joseph Last from Malawi to Quelimane, 1887 ......................... 15 

Fig. 6. Map of Zambezi Company territory (Maugham 1910) ................................................ 16 

Fig. 7. Base camp at abandoned tea estate manager's house, Cha Madal ................................ 17 

Fig. 8. Soviet map of Mt Mabu and Tacuane area, 1:250,000 scale ........................................ 18 

Fig. 9. Supervised classification of Mt Mabu forest vegetation, 2011 (J. Bayliss) .................. 21 

Fig. 10. Partially-supervised classification of Mt Mabu forest vegetation, 2008 (S. Baena) .. 21 

Fig. 11. Moist woodland (Syzygium cordatum) at forest margin ............................................. 23 

Fig. 12. Burnt woodland in agricultural zone on lower SE slopes of Mt Mabu; woodland at 

forest margin showing boundary ................................................................................ 23 

Fig. 13. Medium altitude moist forest, near Mt Mabu campsite .............................................. 25 

Fig. 14. Medium altitude moist forest canopy, Mt Mabu ........................................................ 25 

Fig. 15. Montane forest on upper slopes of Mt Mabu .............................................................. 26 

Fig. 16. View over montane forest on Mt Mabu ...................................................................... 27 

Fig. 17. Coleochloa 'grassland' on rocky slopes near summit of Mt Mabu ............................. 28 

Fig. 18. Upper forest margin with montane shrubland and Coleochloa 'grassland'................. 28 

Fig. 19. Montane shrubland near summit of Mt Mabu ............................................................ 28 

Fig. 20. Montane shrubland near summit of Mt Mabu ............................................................ 28 

Fig. 21. Cryptostephanus vansonii on upper slopes of Mt Mabu ............................................ 34 

Fig. 22. Polystachya songensis orchid on Mabu summit ......................................................... 34 

Fig. 23. Olive Sunbird in Mt Mabu forest ................................................................................ 42 

Fig. 24. Lesser Pouched Rat caught in medium altitude forest, Mt Mabu ............................... 45 

Fig. 25. Two Atheris mabuensis on forest floor ....................................................................... 46 

Fig. 26. Pygmy chameleon, sp. nov. ........................................................................................ 47 

Fig. 27. Butterfly, Papilio ophidicephalus ............................................................................... 48 

Fig. 28. Butterfly, Cymothoe sp. nov. ...................................................................................... 48 

Fig. 29. Butterfly, Baliochila sp. nov. ...................................................................................... 48 

Fig. 30. Woodland clearance for agriculture and fire .............................................................. 52 

Fig. 31. Sun squirrel caught in gin trap, Mt Mabu ................................................................... 53 

Fig. 32. Small mammal trap and fence, Mt Mabu .................................................................... 53 

Fig. 33. Small but powerful gin trap, Mabu summit ................................................................ 54 

Fig. 34. Large gin trap with Hassam Patel, Mabu forest .......................................................... 54 

Fig. 35. Villagers at Mt Mabu forest camp .............................................................................. 55


SUMMARY 

Located in north-central Mozambique, 95 km south-east of Mt Mulanje in southern Malawi, 

Mt Mabu is a granitic massif rising to 1700 m. Much of it is covered in exceptionally welldeveloped 

and little-disturbed moist forest. It was first explored biologically in 2005 as part of 

a Kew-led project funded through the UK's Darwin Initiative, followed by a large expedition 

under the same project in 2008. Initially done with the help of Google Earth imagery, 

discovery of the extensive forest and the subsequent expedition gave rise to much 

international media coverage, some of which is outlined here. This report describes the area 

and its history, and outlines the main biological findings made under the project. 

Covering an estimated area of 7880 ha, with around 5270 ha of this at medium altitude 

(1000‒1400 m), the forests on Mt Mabu are some of the most extensive of this type in 

southern Africa. Mid-altitude forests are now increasingly rare as so many have been cleared 

for agriculture. In addition, Mabu's forest is also remarkably little disturbed, probably as it is 

mostly found on very steep and rugged terrain. Moist woodland (Syzygium cordatum, 

Pterocarpus angolensis, Xylopia aethiopica) and gully forest (Albizia adianthifolia, 

Erythrophleum suaveolens, Newtonia buchananii) are found below 1000 m, along with 

abandoned tea estates on the southern and south-eastern slopes. The main area of mid-altitude 

moist forest, perhaps the biologically richest and most important area, is characterised by 

large trees 40‒50 m high of Strombosia scheffleri, Newtonia buchananii, Chrysophyllum 

gorungosanum and Maranthes goetzeniana, with extensive clumps of the bamboo 

Oreobambos buchwaldii in gullies. Much moister Afromontane forest, 20‒25 m high, is 

found above 1350‒1400 m altitude, typically with Olea capensis, Rapanea melanophloeos, 

Aphloia theiformis, Faurea racemosa and Podocarpus latifolius. Emerging from the 

Afromontane forest at 1600‒1700 m are relatively small areas of granitic (syenite) rock with 

patches of montane shrubland and clumps of the sedge Coleochloa setifera, a common habitat 

in this region of inselbergs. 

Surveys carried out under the Darwin project yielded 249 plant species above 800 m 

altitude, of which two (a mistletoe Helixanthera schizocalyx and the shrub Vepris sp. nov.) 

are new to science and an additional 11 are significant range extensions from the 

Chimanimani Mountains or Tanzania's Southern Highlands and/or new records for 

Mozambique. 

Zoological findings of vertebrates were even more impressive with 126 bird species being 

recorded, including the discovery of significant populations of Cholo Alethe and the race 

belcheri of Green Barbet, along with smaller populations of Dapple-throat, Spotted Ground 

Thrush, Namuli Apalis (previously believed to be endemic to Mt Namuli) and Swynnerton's 

Robin. Of the 12 bat species recorded, one is new (Rhinolophus mabuensis) and two are new 

records for the country. A total of 15 reptile and 7 amphibian species were found, including 

three new species (the Mabu forest viper Atheris mabuensis, a chameleon Nadzikambia 

baylisii and a pygmy chameleon Rhampholeon). Two other snakes and a frog may also be 

new to science.


Much effort went into surveying the butterflies, with 203 species on the checklist 

including 39 new country records. Four of these are new species (Baliochila sp. nov., 

Cymothoe sp. nov., Epamera sp. nov., Leptomyrina sp. nov.) with a further three new 

subspecies. 

At present the threats to the forests of Mt Mabu are not particularly great, with the great 

majority of it showing very little sign of human disturbance apart from hunting/snaring for 

bushmeat. However, there is a notable impact on the forest‒woodland margin from fire and 

clearance for subsistence agriculture. Logging and clearance for commercial agriculture 

remain potential threats, especially with the recent marked expansion of agriculture in 

Mozambique and new ownership of the abandoned tea estates. At present the Mabu area is 

not formally protected, although there are moves to gazette it at a Provincial level and develop 

a conservation project through a local NGO, Justica Ambiental, and Fauna and Flora 

International. 

Twelve recommendations for both management and further research are given. At this 

stage it is not our technical or scientific knowledge on Mabu's biodiversity that is the limiting 

factor for focussed conservation action, but the development of management plans and 

implementation of such action on the ground. As with many forests, the forests on Mt Mabu 

are still able to regenerate and maintain themselves if the underlying ecological factors 

(rainfall, soils, regeneration microclimate and availability of propagules) remain functional. 

And on Mt Mabu these attributes are, at present, still healthy.


1. INTRODUCTION 

The discovery in 2005 of a large expanse of forest on Mt Mabu in north-central Mozambique 

and the subsequent 2008 expedition, gave rise to much interest nationally and internationally 

‒ some said a "lost Eden", others called it the "Google Forest" as it was first noted using 

Google Earth imagery. In truth, the mountain had already been mapped and named and the 

forest was, of course, known to the local population. It is just that early explorers and, more 

recently, scientists and similar professionals had not recorded or visited it. But the large extent 

and relatively little-disturbed nature of the forest was indeed a very significant find, as were 

the number of new species found. 

Lying just west of the district centre of Tacuane in Zambézia Province, about 80 km southeast 

of the Malawi border near Mt Mulanje, Mt Mabu is one of a series of granitic blocks or 

inselbergs rising above the surrounding coastal lowlands. The country is sufficiently rugged 

that it has attracted little agriculture except on the more gentle footslopes, where tea and other 

plantations were established during the Portuguese colonial period. 

Mt Mabu and its forests had no formal protection, nor was it in any way recognised as an area 

worthy of conservation before this project's major expedition in October 2008. It is hoped that 

its now greatly increased profile will lead to sustainable conservation initiatives. 

1.1 Background 

Under a collaborative project – "Monitoring and Managing Biodiversity Loss on South-East 

Africa's Montane Ecosystems" funded by a UK Government Darwin Initiative grant – various 

trips were made to Mt Mabu area from 2005 to 2010 (Table 1), in particular a major 

expedition was undertaken in October 2008 (see Bayliss 2009). This was followed 

subsequently by a smaller trips in 2009‒2010 looking at butterflies and reptiles and for natural 

history filming. Most of these trips were a collaborative effort between the Royal Botanic 

Gardens Kew (RBG Kew), the Instituto de Investigação Agraria de Moçambique (IIAM), the 

Maputo Natural History Museum (MHN), the Mulanje Mountain Conservation Trust 

(MMCT), the Forest Research Institute of Malawi (FRIM), the African Butterfly Research 

Institute (ABRI) in Kenya, and BirdLife International. A full list of participants on the main 

2008 expedition is given in Annex 1. Additional persons who contributed to sections of this 

report are listed in the Acknowledgements. 

The objectives of the study and expeditions were: 

1. To undertake botanical and vegetation field survey of Mt Mabu, 

2. To gather additional zoological information on the mountain, particularly on birds, 

reptiles and butterflies, 

3. To train a team of Mozambican and Malawian biologists in botanical and vegetation 

survey techniques, 

4. To assess the extent, status and threats to the moist forest and other biodiversity on the 

mountain, 

5. Based on gathered field data, to develop species and habitat recovery plans. 

This report presents and discusses findings from the main 2008 expedition, reconnaissance 

and subsequent trips, but also attempts to place these in a broader context as well as draw 

conclusions relevant to Mabu's conservation. Detailed species lists for plants, birds and 

butterflies are presented, along with partial data on bats, reptiles and other groups. In addition


to the species lists we give the first detailed account of the vegetation, and discuss the threats 

to all biodiversity. Particular attention has been paid to endemic, rare or threatened species. 

Our main attention was given to areas and species above 800 m altitude, as below this height 

much of the vegetation has already been transformed. 

Table 1. Main visits to Mt Mabu, 2005‒2010. 

dates taxa recorded persons involved 

Dec 2005 birds, plants, herps, butterflies J. Bayliss, C. Spottiswoode, E. 

Herrmann, H. Patel 

Jan 2006 butterflies, herps J. Bayliss and others 

Sept-Oct 2008 herps, plants, butterflies J. Bayliss, H. Patel and others 

Oct 2008 plants, birds, herps, small main Darwin project expedition 

mammals, butterflies 

May 2009 

Oct-Nov 2010 

butterflies, herps 

butterflies, herps 

J. Bayliss, W. Branch, ABRI 

and others 

J. Bayliss, ABRI, FFI 

Nov-Dec 2010 butterflies, herps 

J. Bayliss, BBC filming 

1.2 Media Coverage 

Unlike other expeditions to mountains in northern Mozambique and Malawi carried out under 

this Darwin Initiative project, the 2008 trip to Mt Mabu gave rise to a phenomenal amount of 

international media attention. This coverage and interest ‒ which still continues in 2012 ‒ is 

outlined in Annex 2, along with a selection of references to some of the main articles and web 

links, including videos.


2. DESCRIPTION OF STUDY AREA 

2.1 Geography and Geology 

The highest of a series of blocks, Mt Mabu rises above the surrounding lowland plains at 

around 350‒450 m altitude, just north of the Rio Lugela in Zambézia Province, north-central 

Mozambique. It is centred on 16 o 17'S, 36 o 24'E, with the 1710 m summit at 16 o 17'56.5"S, 

36 o 23'44.3"E. Situated within the District of Lugela, it lies some 95 km south-east of Mt 

Mulanje in southern Malawi, 120 km south-west of Mt Namuli and 200 km from the 

Provincial Capital of Quelimane on the Indian Ocean coast (Figure 1). The district centre of 

Lugela is 40 km away, while the larger town of Mocuba, at the confluence of the Lugela and 

Licungo rivers, is 85 km to the south-east. 

The Mabu massif is essentially a complex of granitic inselbergs ('whalebacks') or ancient 

igneous intrusions, exposed by millions of years of subsequent erosion. It is significantly 

smaller than the Namuli complex (see Timberlake et al. 2009) and, unlike that massif, does 

not include any substantive area of upland plateau. The rock forming the Mabu massif is 

syenite, similar to granite, an igneous intrusion of the younger Precambrian Namarroi series 

dating from 850‒1100 Mya (Instituto Nacional de Geologia 1987). This intrusion is 

surrounded by migmatites, also of the Namarroi series. 

Lugela District contains four Posto Administrativos and 16 Localidades ‒ Mabu is in the 

Posto Administrativo (P.A.) of Tacuane, Mabu Localidade. The nearest main administrative 

centre is Tacuane 15 km away (Figure 2), with the lesser administrative post of Limbuè lying 

on the southern footslopes of Mabu. The population of P.A. Tacuane in January 2005 was 

18,191 persons (Ministério da Administração Estatal 2005), a relatively low number. 

Population pressure is not high in this area, although the area may well have been more 

populous during the colonial period when the tea estates were providing employment. Areas 

visited by the 2008 expedition overlain by contours are shown in Figure 3. 

A GIS-based altitudinal analysis of the Mt Mabu area by J. Bayliss showed that out of a broad 

study area surrounding the Mabu massif of around 300 km 2 , 8308 ha lies above 1000 m 

(Table 2), which is around the lower limit of true moist forest (at least on the eastern slopes), 

although lowland gully forest can be found below this (see Section 4.3.3). 

Table 2. Extent of area above 1000 m by altitudinal class 

for the Mt Mabu massif. 

Altitude (m) extent (ha) % 

1000‒1200 3527 42.5 

1200‒1400 3723 44.8 

1400‒1600 1031 12.4 

1600+ 27 0.3 

Total 8308 100.0 

Although the forest on Mt Mabu was first noted on Google Earth (Figure 4a), and has 

sometimes been called the "Google Forest", there are some inaccuracies present on the 

webpage images (accessed January 2012). The main one is that the peak of Mt Mabu is 

shown on Google Earth as being 18 km east of its actual location ‒ the peak shown as "Mt 

Mabu" is a wooded area away from the main massif. Secondly, on the ground it is clear that 

the peak is along a bare open ridge south-west of the main rock outcrop, but the highest point 

appearing on Google Earth imagery is 1651 m, not 1710 m as on all cadastral maps, and is


apparently situated on a vegetated lower ledge some 600 m to the south-east. This could be an 

artefact resulting from cloud cover. 

Fig. 1. Montane areas in N Mozambique studied 

under the Darwin Initiative project [JB]. 

Fig. 2. Google Earth image of Tacuane area showing Mt Mabu (centre left) and roads (Aug 2006)[JF]. 

2.2 Climate 

Climate data from the Mabu massif itself above 1000 m altitude are not available, but data for 

the Madal tea estates near Tacuane (16 o 21'S, 36 o 22'E, 400 m altitude), possibly at Limbuè just


7 km away, is summarised in Kassam et al. (1981). These data are for only 16 years and 

probably date from the mid-1960s. 

Mean annual rainfall is given as 2119.1 mm, ranging from a monthly mean of 34.2 mm in 

September to 362.3 mm in January. The main rainfall months are November to April (1793.1 

mm over 6 months or 84.6% of annual total), while the four months from December to March 

have a mean of 1410.9 mm (66.6% of total). Over the 16 years recorded the wettest months 

were March (mean 381.1 mm) and January (mean 362.3 mm). 

Mean annual temperature was 23.7 o C, ranging from 21.0 in July to 25.5 o C in October. The 

mean maximum of 32.9 o C was in October with a mean minimum of 14.9 o C in July. Unlike 

the situation on Mt Namuli (Timberlake et al. 2009), the occurrence of frost is likely to be 

rare. Evapotranspiration (Penman) was 1252.6 mm/year ranging from 63.7 mm in June to 

142.5 in October. During the cooler winter months potential evapotranspiration is roughly 

equivalent to rainfall, but in October it is more than three times monthly rainfall. 

According to Reddy (1984) in his overview of Mozambique's climate, rainfall in the area 

should be around 1500 mm/year (surprisingly less than shown by actual rainfall records) with 

a low variation of only 20%. The winter rainfall index (30) indicates that some winter 

cropping is possible. In a national context the Cha Madal area is a relatively high rainfall area 

similar to Mt Namuli (zone 1‒2a, moderately cool; national climatic resources inventory, 

Voortman & Spiers 1982), with the possibility of two rainfed growing periods in 10% of 

years, perhaps with a 300 day growing period each year. 

Fig. 3. Map of Mt Mabu area showing forest extent and main localities visited (Oct 2008)[JB].


2.3 Aerial Photos 

The only aerial photos apparently available are from August 1965 and June 1969 at a given 

scale of around 1:43,000, although subsequent analysis and measurements from the 1:50,000 

map sheet suggest the scale is actually around 1:35,000. Owing to distortion, all area 

determinations were made using satellite imagery. 

Aerial photos of Mt Mabu, ±1:43,000 scale: 

row 418/062‒072 (north), 17 August 1965 

row 3111/076‒086 (mid), 15 June 1969 

row 418/118‒111 (south), 17 August 1965 

Fig. 4a,b. Google Earth image of Mt Mabu massif from 2006 (left) showing forest + dense woodland 

cover in dense green and abandoned tea plantations in paler blue-green (lower right and lower centre, 

with straight lines) and false-colour Landsat image (right) from July 2005 with dense vegetation cover 

shown in red.


3. HISTORY AND EARLY COLLECTING 

3.1 Early History 

We have not been able to find any reports of early colonial explorers noting or visiting the 

Mabu massif, at least in the British sources that describe the exploration of Mt Namuli and 

travels across northern Mozambique to southern Malawi in the latter 19 th and first half of the 

20 th century (e.g. Johnson 1884, O'Neill 1884, Last 1887, Vincent 1933a). 

Joseph Last, an intrepid 19th Century British traveller, carried out a lengthy journey in 1885‒ 

1887 from the southern Tanzania coast to Blantyre in what is now Malawi, around parts of 

southern Malawi and then down to the coast at Quelimane and back to Blantyre (Last 1887, 

1890, Timberlake et al. 2009). Part of this long trip involved a foot safari from Mt Namuli 

down along the Rio Licungo to the coast and then, in mid-December 1886 after a few days 

rest in Quelimane, upstream along the Rio Licungo to what is now Mocuba, along the Rio 

Lugela and through the hilly country north of Lugela town to what is now Namarroi. From 

there he walked along the Rio Luo entering Malawi just south of Lake Chilwa (Chirwa) and 

north of Mt Mulanje, arriving in Blantyre on 14 January 1887. Unfortunately, in the account 

published by his sponsors, the Royal Geographical Society (Last 1887), he hardly mentions 

the Lugela leg of this trip. The route he took is shown on a subsequent map (Last 1890, 

portion shown in Figure 5). However, despite attempts to match the Lugela portion with 

present 1:250,000 and 1:50,000 map sheets, his route could not be traced except in fairly 

broad terms. 

All that can be said is that Last left the Rio Lugela heading north upstream of present-day 

Mocuba (not marked on his map) somewhere between Mt Cuba/Murra (16 o 40'S, 36 o 50'E) and 

just upstream of the Xilusi‒Lugela confluence (16 o 32'S, 36 o 38'E). From there he made his 

way between the hills west of Mt Mavigue ("Mavugwe", 16 o 22'S, 36 o 45'E), passing east of 

what is now Muabana but west of Mt Muiane ("Mwiani",16 o 02'S, 36 o 37'E), and then crossing 

the present-day Rio Mucodi soon after, some 30 km west of Namarroi, before travelling along 

the Rio Luo. Thereafter he followed the Luo westwards towards Malawi, which he entered 

just south of Lake Chilwa (Chirwa). 

In those days, fixing longitude was not easily done and distances were estimated from time 

spent walking, so many distortions are apparent in the 1890 map (Figure 5). Thus it has not 

been possible to identify more than a very few of the hills or other points it shows, and the 

relative positions of peaks seem highly distorted compared to present-day maps. However, 

although it seems fairly certain that of the early recorded travellers Last perhaps came closest, 

he did not pass close to Mt Mabu or see it (despite an intriguingly-named peak "Mapu 

H"[hill] on his 1890 map), or travel along the route of the present-day Tacuane‒Muabana 

road. Unfortunately, in his account there is also no mention of the type of country he passed 

through, the people or the life in these areas, or even whom he went with. 

Earlier, in 1884, Daniel Rankin and H.E. O'Neill (then the British Consul on Mozambique 

Island) plus 23 porters, walked from Blantyre to Quelimane along an existing and significant 

trade route to see how quick and effective that route was to the coast (Rankin 1885). It took 

them just 39 days. At the time there was concern that the Rio Zambezi was becoming less 

passable to boats owing to river silting, and that this route into the important new settlements 

of Blantyre and Zomba in southern Malawi could be easily cut by hostile forces. Their route 

passed south of the Mulanje massif, across the headwaters of the Rio Lugela but kept to 

higher ground until it met the Rio Tetema. From here it followed the river down to the 

Munguze ("Mulongusi") confluence and then down the Rio Licuare ("Likwati") to 

Quelimane. However, this route, also shown on Last's map (Last 1990), passes quite a long 

way south of Mt Mabu.


Fig. 5. Map showing route of Joseph 

Last from Malawi to Quelimane, 

1887. 

Another book by an early traveller (Maugham 1910) on the whole Zambezia area does not 

mention Mabu or the area around, focussing more on the coastal strip. Although the Namuli 

area is clearly shown on his map, the Mabu area is almost omitted (Figure 6). 

Hence despite a significant amount of trading and travelling in this part of north-central 

Mozambique, it appears that the Mabu area was continually by-passed, perhaps because of its 

rugged terrain, poor agricultural potential, thus scattered and low human population.


Fig. 6. Map of Zambezi Company territory (from Maugham 1910). 

3.2 Recent History 

The first major economic development in the area appears to have been the establishment of 

tea plantations in what was then Tacuane District on the lower slopes of Mt Mabu in the 

1930s (Wilson, Smithett & Co. 1962). By 1961 there were three tea estates in the area and 

two tea factories, compared to 14 estates and 11 factories in the Gurué area around Mt Namuli 

and 7 estates and 3 factories in the Milange area just across the border from Mt Mulanje in 

Malawi. Details of the three Tacuane estates are given in Table 3. From these records it 

appears that the estate where the present Darwin expedition was initially based ‒ Cha Madal ‒ 

was by far the largest. 

Table 3. Tea enterprises in the Mt Mabu area, with 1960 production figures 

(from Wilson, Smithett & Co. 1962). 

Company 

altitude 

(m) 

Company 

area under 

tea (ha) 

tea production 

(kg) 

Cha Madal 400 Sociedade Agricola do Madal 607.5 444,466 

Cha Tacuane 700 Manuel Nunes 342.2 182,217 

Cha Lugela 700 João Martins 93.2 (no factory) 

1042.9 626,683 

The area under tea in northern Mozambique in the 1960s was surprisingly large at 15,010 ha, 

greater than that in Malawi (12,110 ha) or Kenya (14,950 ha). Mozambique's total tea 

production in 1961 was 23,368,000 lbs [pounds] (10,609,072 kg), with exports being 

21,362,000 lbs (9,698,348 kg), compared to production of 31,518,000 lbs (14,309,172 kg) in 

Nyasaland [Malawi], 27,869,000 lbs in Kenya, 9,830,000 lbs in Tanganyika [Tanzania] and 

2,379,000 lbs in Southern Rhodesia [Zimbabwe]. Hence production per hectare in 

Mozambique at 706.9 kg/ha was significantly less than that in Malawi (1182 kg/ha) or Kenya 

(847 kg/ha) (all figures from Wilson, Smithett & Co. 1962). Tea production was obviously a


major economic and agricultural activity in the area in the 1960s, but probably declined 

significantly in the 1970s with the Independence struggle and subsequent civil war. 

At the base of Mt Mabu on the southeastern side is the now-derelict Cha Madal Tea Estate, 

until recently owned by Madal (now Rift Valley Holdings), a large Mozambican company 

best known for its coconut plantations on the coast near Quelimane. In 1961 the estate 

manager was John Edge (Wilson, Smithett & Co. 1962). It was obviously a large estate, with 

an extensive factory, out-buildings, workers' and managers' housing. What was presumably 

the Estate Manager's house (16°18'20.5"S, 36°25'28.8"E, Figure 7a,b) was a large, spreading, 

opulent building with a wonderful view over the surrounding plains. The estate was 

abandoned in August 1982 as fighting in the area reached a stage where tea production was no 

longer viable; the tea estate later became a Renamo base. 

Fig. 7a,b. Base camp at abandoned tea estate manager's house, Cha Madal (left TH, right CS). 

Around the Estate Manager's house many exotic tree species were planted, presumably both 

for utilitarian purposes and for shade and beautification. These included Artocarpus 

heterophylla (Jackfruit), Eucalyptus cf. grandis (probably for fuelwood and construction), 

Vernicia montana (Tung oil), Grevillea robusta, Delonix regia, Ceiba pentandra (Kapok), 

Encephalartos cycads and Ficus elastica (India Rubber tree). 

The area of plantation tea can be clearly seen in the satellite imagery as a vegetation block 

south-east of the main forest area (Figure 4a). The estate was planted with over 600 ha of 

China hybrid tea (Camellia sinensis cultivar.), of which 520 ha were in production as of June 

1961 (Wilson, Smithett & Co. 1962). As it has not been managed for almost 30 years, the tea 

bushes have grown up to 15 m forming an almost solid-canopy forest. Apparently, it may still 

be commercially viable if pruned, and could be linked to neighbouring tea-producing areas in 

Gurué or even in Mulanje in southern Malawi. Recently, a senior Madal representative 

suggested that the planned rehabilitation of the estate may take the form of removing the tea 

and replanting with other, more viable crops. However, Madal have now (2011 or early 2012) 

sold the estate to a company called Mozambique Holdings. 

There is a Soviet map of the area, possibly forming part of a national 1:250,000 series 

produced in the late 1970s or early 1980s. The section of the sheet covering Mabu is shown in 

Figure 8. It gives a bit more detail than is available on the national Dinegeca 1:250,000 scale 

map series, and seems to have been derived separately, possibly from more recent aerial 

photos. 

On the summit of Mt Mabu we noted the destroyed remains of a trigonometrical point 

(presumably from which the spot height of 1710 m was determined), and this was


subsequently borne out by David Scott (pers. comm., Feb. 2009) who reported constructing 

such a point with the Mozambique mapping agency Dinegeca in 1995. 

Fig. 8. Soviet map of Mt Mabu and Tacuane area, 1:250,000 scale (probably 

late 1970s). Mt Mabu indicated with red arrow.


4. VEGETATION 

4.1 Previous Studies 

It seems that there have been no previous studies on, or even recognition of, the vegetation of 

Mt Mabu or the immediate area. However, as mentioned earlier, the presence of a number of 

tea estates on the lower slopes means that the surrounding vegetation must have been known 

to agriculturalists, even if it appears not to have been formally documented. 

On a continental or regional scale, the Mabu area is shown by White (1983) as forest patches 

of the East African coastal mosaic (type 16b) surrounded by Wetter Zambezian miombo 

woodland (type 21). White is wrong in his assertion that these forests are linked to those on 

the East African coast as they are clearly montane and medium altitude forests, very similar to 

those found on mountains in southern Malawi and Eastern Zimbabwe. The more detailed 

study by Wild & Barbosa (1968), on which White's study for this area was based, maps the 

Mabu area as Moist Evergreen Forest at low and medium altitudes (Type 1) surrounded by 

Brachystegia spiciformis (high rainfall) woodland (Type 21). 

The more detailed (and earlier) map of Zambézia Province by Barbosa (1952) shows the 

Mabu massif as Unit 1 "Floresta higrófila tropical altimontana, de chuvas e nevoeiros" (moist 

high-altitude rain and cloud forest), the same as Mt Namuli, surrounded by Unit 2 "Floresta 

sub-higrófila, das altitudes medias, de Brachystegia spiciformis com elementos da floresta 

higrófila" (medium-altitude sub-moist forest [woodland] with Brachystegia spiciformis and 

moist forest patches). They describe the moist status due to incoming rain and clouds, with a 

transition to xerophytic cold-adapted vegetation higher up. A few forest species are listed. 

A subsequent, more detailed study by Pedro & Barbosa (1955) looked at vegetation across the 

whole country from an agro-ecological viewpoint. The accompanying map shows Mabu 

("Alto Lugela") as unit 79 (Zonas altimontanas da Zambézia‒Niassa) occurring between 

1000‒1800 m. However, they mention that these zones have not been visited by them so they 

can not say much, and no species are listed. 

4.2 Vegetation Mapping 

Vegetation description of Mt Mabu was carried out by us in two ways ‒ determination of the 

possible extent of forest using satellite imagery supported by the use of historical 

panchromatic aerial photos, and categorisation of vegetation types seen up the altitudinal 

gradient to the peak in the south-eastern part of the massif. 

Two separate studies were done for area determination. One was carried out in 2006 using 

Landsat 7 ETM+ image (reference S-37-15-2000, 30 m resolution) from the year 2000 viewed 

through very near infra-red (VNIR) filters (Bayliss in Spottiswoode et al. 2008). This was not 

a supervised classification. The second area determination was done by Julian Bayliss in 2011 

using a supervised classification based on locations noted in the field during October 2008 

(see below). 

Results from the first, Spottiswoode et al. study, although not rigorous, suggested an area of 

dense vegetation above 1000 m altitude ‒ which was assumed to be moist forest, but this had 

not been confirmed at that stage ‒ of between 5000 and 7000 ha, excluding the fairly obvious 

extent of old tea plantation (about 2000 ha, seen through the presence of straight margins, 

Figure 4a).


The second determination of forest extent (J. Bayliss) ‒ the one which we are using here ‒ 

was done by making a draft vegetation map based on an unsupervised classification using 

Erdas Imagine (maximum likelihood algorithm applied to a 6-band stack image) of a Landsat 

ETM+ image with 30 m resolution from July 2005 (path 166, row 071). Twelve classes in 9 

broad habitat types were recognised, including an 'unclassified' class. 

Following fieldwork a final vegetation map was developed using a supervised classification 

of the same Landsat image with radiometric and geometric correction, in which four broad 

habitat types were separated out ‒ moist forest, woodland, agriculture, rock and bare ground 

(Figure 9). Based on this latter interpretation, it was calculated that 6937.4 ha of moist forest 

were present on Mt Mabu, most of it above 1000 m; a substantial buffer of woodland is also 

shown. Although the figure for forest extent may be an overestimate as the difference between 

moist forest and dense woodland is not clear-cut and shadow effects may be significant, it is 

thought to be the best estimate possible at this stage without more detailed interpretation and 

field survey. Using these figures, forest covers 66% of the total area above 1000 m (see Table 

2). 

This forested area was divided into altitudinal classes (see Table 4). Out of a total 

(planimetric) forest area of 6937.4 ha, 4563.6 ha lies between 1000 and 1400 m, which we 

consider to be primarily mid-altitude moist forest, and an additional 919.5 ha lies above 1400 

m, which we consider to be high altitude or Afromontane moist forest. In addition, there is a 

significant amount of forest below 1000 m, but much of this is riverine or gully forest and 

perhaps some is overgrown plantation. The main forest block is that area above 1000 m 

altitude, which is 5483 ha. 

Table 4. Extent of forest area by altitudinal class for the Mt Mabu massif 

(derived from J. Bayliss supervised classification, 2011). 

Altitude (m) 

measured planimetric 

extent (ha) 

% estimated extent using 

slope correction factors (ha) 

below 1000 1454.3 21.0 1600 

1000‒1200 1719.9 24.8 

1200‒1400 2843.7 41.0 

5270 

1400+ 919.5 13.2 1010 

Total 6937.4 100.0 7880 

What should be also taken into account is that these forests are mostly on steep slopes, hence 

the actual area covered is significantly larger than the planimetric figures given above. Using 

an approximation of a 30 o slope between 1000‒1400 m (not an unreasonable assumption 

given the steep slopes there) and an estimated 15 o slope below 1000 m and above 1400 m, 

coupled with tangent tables, a rough rounded estimate of forest area in these various 

altitudinal bands is shown in the right-hand column of Table 4. The forest extent in the 

1000‒1400 m band is 5270 ha, higher than the cumulative planimetric figure of 4564 ha. 

Coupled with the forest extent above and below on less steep slopes, this gives a total forest 

cover on the mountain (excluding the tea plantations) of around 7880 ha. 

In totally separate exercises, the National Land Cover map of Mozambique gives the total 

extent (planimetric) of forest cover in the Mabu area as around 5500 ha (J. Francisco, pers. 

comm. 2010), while Susana Baena (RBG Kew GIS Unit) did an initial partially supervised 

classification using 26 ground control points derived from a reconnaissance in June 2008 that 

arrived at a more conservative figure of 5998 ha (see Figure 10).


All these studies show that the area of moist forest is very extensive for the region (between 

5500 and 7900 ha) with the great majority of it being found between 1000 and 1400 m. Such 

mid-altitude forest is increasingly rare in the southern African region as these areas have often 

been cleared in the past 100 years for timber and agriculture. We believe that it represents 

perhaps the largest extent of moist forest at such altitudes in southern Africa; in addition these 

forests are in excellent condition and little disturbed. 

Fig. 9. Supervised classification of Mt Mabu forest vegetation, 2011 (J. Bayliss). Red dots 

indicate routes travelled on 2008 expedition. 

Fig. 10. Partially-supervised classification of Mt Mabu forest vegetation, early 2008 

(S. Baena, RBG Kew). Red dots indicate path used in 2006.


4.3 Vegetation Types 

As the expedition only studied vegetation of the lower slopes in the south-eastern corner 

above the abandoned tea estate, and was not able to visit the majority of the forested area of 

Mt Mabu, any account of the vegetation must be limited in scope. The account below covers 

primarily the south-eastern side of the mountain that was extensively visited. Vegetation 

types and patterns on the drier western and northern slopes appear to be somewhat different 

from those described here (see Section 4.3.5). 

Above 600 m altitude vegetation on the Mabu massif can be classified into three main groups 

‒ woodland, forest, and scrub/sedge patches on bare rock. Below this altitude abandoned 

plantations and secondary vegetation are found, along with patches of lowland riparian forest. 

Above 800 m the forest group can be further subdivided into tall riparian forest, tall medium 

altitude forest and shorter high altitude ("montane") forest (sub-montane in the classification 

of Müller 1999), but within which there was significant variation. 

The great majority of the area studied was under forest, the main vegetation type of interest. 

In our study less emphasis was placed on woodland and the vegetation of rocky outcrops 

around the summit. Boundaries between the different vegetation types were sometimes 

surprisingly clear-cut, e.g. between montane forest and low scrub on the summit, and in 

places between woodland and medium altitude forest. These 'hard' boundaries may in part be 

due to fire. 

The main vegetation types are characterised and described below in terms of their structure 

(height, cover, etc.), species composition and ecology. This was done both through general 

observation and through the recording of 48 vegetation survey plots, each of around 0.25 ha 

extent (forest) to 0.5 ha (more open habitats), placed subjectively in what were considered to 

be representative sites. A GPS point was recorded for each. Additional clarification was 

obtained from viewing old aerial photographs (see Section 2.3) and the use of 0.04 ha forest 

mensuration plots (see Section 4.4). The descriptions follow an altitudinal sequence. 

Additional details on vegetation are given in Dowsett-Lemaire and Dowsett (2009). 

4.3.1 Plantations (400‒600 m) 

Around the ruined tea estate manager's house (16°18'20.5"S, 36°25'28.8"E, 550 m) there are a 

number of overgrown plantations of tea (Camellia sinensis) and Eucalyptus cf. grandis. Tea 

bushes, originally kept to around 1.5 m in height, have now grown up to 12–14 m high, many 

2–3 stemmed with stems 10–15 cm in diameter. These are overtopped by Albizia 

adianthifolia, locally forming a closed canopy. Other native trees found here include 

Macaranga capensis. 

Much of the planted and secondary growth around this area contains exotics (Grevillea 

robusta, Delonix regia, Eucalyptus cf. grandis and Vernicia montana) mixed with pioneer 

forest trees such as Macaranga capensis. Other exotics planted near the estate manager's 

house include Ceiba pentandra, Artocarpus heterophyllus (jackfruit) and Ficus lutea. 

4.3.2 Woodland (600‒1000 m) 

This type was not examined in detail, but on the lower sections of the main path up to Mt 

Mabu the woodland is clearly dominated by Pterocarpus angolensis. Other common trees 

include Pteleopsis myrtifolia and Vitex doniana, while Pericopsis angolensis and 

Stereospermum kunthianum were noted occasionally. The woodland is underlain by a carpet 

of Aframomum albiflorum; clumps of Oxytenanthera abyssinica bamboo often grow at the 

ecotone between woodland and dry forest or near dry streams. With increasing altitude, 

Syzygium cordatum becomes more common until it is locally dominant above 800 m (Figure 

11), forming pure stands sometimes closed enough to be called forest. Aframomum remains


common under Syzygium and the number of low-level epiphytes (ferns and orchids, also 

Rhipsalis) at 800–950 m suggests a high level of humidity for much of the year. 

At 900 m the transitional woodland‒forest on the ridge is drier, being dominated by Syzygium 

cordatum and Xylopia aethiopica. Other species noted were emergent Newtonia buchananii, 

Albizia adianthifolia and Macaranga capensis, with an understorey of Craterispermum 

schweinfurthii, Cussonia arborea, Englerophytum magalismontanum, Erythroxylum 

emarginatum, Oxyanthus speciosus, Phoenix reclinata, Synsepalum cerasiferum and 

Tabernaemontana ventricosa. Lianas were mainly Dalbergia lactea, Landolphia kirkii and 

Urera trinervis. 

Fig. 11. Moist woodland (Syzygium 

cordatum) at forest margin (JT). 

Fig. 12 a,b. Burnt woodland in agricultural zone on lower SE slopes of Mt Mabu (left, TH); woodland 

at forest margin showing boundary (right, JT). 

4.3.3 Moist Forest (400‒1650 m) 

The following categories of forest can be recognized: lowland riparian forest (400–900 m), 

mid-altitude moist forest (980 to 1350–1400 m) and Afromontane moist forest (from 1350– 

1400 m to 1650 m). Only the latter two were extensively studied by us. 

a) Lowland riparian forest (400‒1000 m) 

This forest type occurs over a significant altitudinal range and is fairly narrow in extent, so 

varies greatly in its composition and structure. 

Lower down, in patches of lowland riparian forest at 400–500 m near the tea estate 

managers's house (16°18'26"S, 36°25'39"E), large (40‒50 m high) trees of Albizia 

adianthifolia (dominant), Erythrophleum suaveolens, Khaya anthotheca, Macaranga


capensis, Newtonia buchananii, Parkia filicoidea, Pteleopsis myrtifolia and Synsepalum 

cerasiferum were noted in the canopy. Edge or pioneer species included Bridelia micrantha, 

Harungana madagascariensis, Trema orientalis and Vitex doniana, while understorey species 

included Dracaena mannii, Celtis gomphophylla, Clausena anisata, Ensete ventricosum and 

various lianes. 

Lowland riparian forest at 800–900 m near the path to the main forest camp was characterised 

by 45 m high emergent trees of Newtonia buchananii (Khaya anthotheca was rarely seen), 

with other canopy trees being Albizia adianthifolia, Anthocleista grandiflora, Erythrophleum 

suaveolens, Macaranga capensis, Parinari excelsa, Synsepalum cerasiferum and Xylopia 

aethiopica, with Pteleopsis myrtifolia, Vitex doniana and, locally, Shirakiopsis (Sapium) 

elliptica at the edges. The commonest understorey species were Craterispermum 

schweinfurthii and Erythroxylum emarginatum; others include Dracaena mannii, Oxyanthus 

speciosus, Englerophytum magalismontanum, Tabernaemontana ventricosa, small trees and 

saplings of Cryptocarya liebertiana, Cussonia spicata and Polyscias fulva (from 850 m). 

Oreobambos buchwaldii bamboo was fairly common, and locally the palm Phoenix reclinata. 

The tree fern Cyathea dregei occurs along the main streams, the large fern Marattia fraxinea 

is frequent, and the shrub Carvalhoa campanulata grows in light gaps. The commonest 

canopy liana by far is Millettia lasiantha, found with various Apocynaceae (Dictyophleba, 

Landolphia kirkii, Saba comorensis), Combretum paniculatum, Dalbergia lactea and Urera 

trinervis. 

b) Mid-altitude moist forest (980‒1400 m) (Figures 13, 14) 

This is found between the altitudes of 980–1000 m and 1350–1400 m, after which a sudden 

change in the dominant canopy species occurs. 

From the forest plots (see Section 4.4) and general observation, the main forest canopy trees 

in terms of basal area in the lower parts of medium-altitude forest were Strombosia scheffleri, 

Newtonia buchananii, Chrysophyllum gorongosanum and Maranthes goetzeniana. In 

addition, Cryptocarya liebertiana, Ficus sansibarica and Trichila dregeana were also seen. 

Occasional large trees of Cassia angolensis were noted elsewhere. Large strangling figs at 

1000–1350 m belong to two species, Ficus sansibarica and Ficus thonningii, replaced at 

higher elevations by Ficus scassellatii. Away from the stream gullies the forest canopy is 

usually closed, except for small gaps caused by tree-falls. 

The main sub-canopy trees (often with more stems but alower basal area) were Drypetes 

gerrardii, Drypetes natalensis, Funtumia africana, Garcinia kingaensis, Rawsonia lucida, 

Tabernaemontana ventricosa and a number of Rubiaceae including Heinsenia diervilleoides, 

Aidia micrantha, Tricalysia acocantheroides (all the way to the top) and Tricalysia pallens. 

Other sub-canopy trees and shrubs seen included Allophylus chaunostachys, Blighia 

unijugata, Cola greenwayi, Diospyros abyssinica (starts appearing just above 1100 m), 

Myrianthus holstii, Oxyanthus speciosus, Vepris nobilis and Zanthoxylum gilletii. 

Haplocoelum foliolosum is common from 1150–1300 m and the first big Tabernaemontana 

stapfiana appear on the ridge at 1200–1250 m. Canopy lianas are dominated by Millettia 

lasiantha, with Acacia pentagona, Agelaea heterophylla, Combretum paniculatum, 

Dictyophleba lucida, Landolphia kirkii, Oncinotis tenuiloba and Urera trinervis also 

common. 

Enormous clumps of Oreobambos buchwaldii bamboo to 18–20 m tall are frequent all the 

way up to 1400 m, particularly on dry slopes and in gullies. 

At around 1000 m near the main forest camp (16°17'10"S, 36°24'01"E), tall trees reach an 

impressive height of 40–45 m (see front cover), with Strombosia scheffleri being the


commonest (largest between 45–50 m tall, 90 cm DBH). Apart from Strombosia, common 

canopy large trees here were Newtonia buchananii (largest seen 50 m tall, 140 cm DBH), 

Chrysophyllum gorungosanum (over 45 m) and Maranthes goetzeniana (40 m, 35 cm DBH). 

At about 1100 m altitude, beyond a large rocky stream (16°17'05"S, 36°23'44"E) on a gentle 

slope, the forest is very impressive with the tallest trees (at least 40 m) including many 

Strombosia. Several Chrysophyllum gorungosanum, Maranthes goetzeniana and Newtonia 

buchananii were noted, with one or two tall Cryptocarya liebertiana, Trichilia dregeana and 

strangling figs (Ficus sansibarica). In the subcanopy Drypetes gerrardii, Garcinia kingaensis 

and Myrianthus holstii were seen. Smaller trees include Drypetes natalensis (also conspicuous 

along the stream, with arching branches), Pavetta gurueënsis, Rawsonia lucida, Rinorea 

ferruginea, Vepris sp. nov. and Synsepalum muelleri. 

Steep, wide gullies with permanent streams contain more light-demanding tree species such 

as Albizia adianthifolia, Macaranga capensis, Newtonia and Polyscias fulva (the latter 

becoming bigger and more frequent with increasing altitude). Also found here were 

Anthocleista grandiflora, Funtumia africana and medium-sized Bridelia micrantha, 

Englerophytum magalismontanum, Xylopia aethopica and small Bersama abyssinica. Tree 

ferns (Cyathea dregei) occur along streams to at least 1400 m while Dracaena fragrans is 

common in humid hollows and on some slopes. 

Fig. 13 a,b. Medium altitude moist forest, near Mt Mabu campsite (JB). 

Fig. 14. Medium altitude moist 

forest canopy, Mt Mabu (JT). 

c) Afromontane forest (1350‒1650 m) (Figures 15, 16) 

This forest type with its lower canopy height and much moister aspect is found up to 1650 m. 

The change from medium-altitude to high altitude forest is fairly abrupt at about 1350‒1400


m, at least on the south-eastern slopes, and is seen in the dropping out of Newtonia 

buchananii, the replacement of Albizia adianthifolia by A. gummifera, and in Olea capensis 

becoming a conspicuous tall tree. 

Canopy trees in the lower parts of Afromontane forest include Strombosia scheffleri, 

Chrysophyllum gorungosanum, Maranthes goetzeniana and Newtonia buchananii, with Cola 

greenwayi, Garcinia kingaensis, Heinsenia diervilleoides, Myrianthus holstii, 

Tabernaemontana stapfiana and Vepris nobilis in the sub-canopy. Small Cassipourea 

malosana and the understorey tree Lasiodiscus usambarensis appear around 1300 m, while 

Maytenus acuminata and Eugenia capensis subsp. nyassensis are common between 1300– 

1400 m. Higher up Podocarpus latifolius becomes increasingly common. Anthocleista 

grandiflora and Polyscias fulva are found in openings or gaps. 

One site on a ridge in the transition zone (16°17'31"S, 36°23'34"E, 1400 m), had several large 

Newtonia, Olea capensis, Parinari excelsa and Polyscias fulva (30–35 m tall), with Aphloia 

theiformis in the gaps. Other trees present included Chrysophyllum gorungosanum, 

Maranthes goetzeniana, Strombosia scheffleri and Zanthoxylum gilletii in the canopy, and 

Cola greenwayi, Craibia brevicaudata, Garcinia kingaensis, Myrianthus holstii, 

Tabernaemontana stapfiana and Vepris nobilis in the subcanopy. Common small trees and 

shrubs were Alchornea hirtella, Heinsenia diervilleoides, Carissa bispinosa, Chassalia 

parvifolia, Clausena anisata, Diospyros abyssinica, Dovyalis macrocalyx, Dracaena 

laxissima, Drypetes natalensis, Erythrococca polyandra, Eugenia capensis subsp. nyassensis, 

Lasianthus kilimandscharicus, Maytenus acuminata, Mostuea brunonis, Pauridiantha 

paucinervis, Pavetta gurueënsis, Peddiea fischeri, Psychotria zombamontana, Rinorea 

angustifolia, Rytigynia uhligii, Synsepalum muelleri, Tricalysia acocantheroides, Vepris sp. 

nov., Vepris nobilis and Memecylon sp. (FD-L 2538, 2–4 m tall). 

At the upper end of the forest at 1600 m, the taller trees (to 25 m high) are Olea capensis and 

Rapanea melanophloeos, with lower trees of Aphloia theiformis, Bersama abyssinica, 

Cassine aethiopica, Cassipourea malosana, Cryptocarya liebertiana, Faurea racemosa, 

Macaranga capensis, Nuxia congesta, Ochna holstii, Pittosporum viridiflorum, Podocarpus 

latifolius, Polyscias fulva, Prunus africana and Syzygium guineense subsp. afromontanum. 

Conspicuous lianas at the forest edge include Rutidea orientalis and Schefflera goetzenii, 

which are already common around 1400 m, and Canthium gueinzii. 

Lower down at 1550‒1600 m the following small understorey trees and shrubs were noted: 

Carissa bispinosa, Chassalia parvifolia, Diospyros abyssinica, Diospyros whyteana (at 

edges), Dovyalis macrocalyx, Dracaena laxissima, Erythroxylum emarginatum, Eugenia 

capensis, Lasianthus kilimandscharicus, Maytenus acuminata, Mostuea brunonis, Pavetta 

gurueënsis, Rinorea angustifolia, Rytigynia uhligii, Tricalysia acokantheroides, Memecylon 

sp. (FD-L 2538) and Vepris nobilis. 

Fig. 15. Montane forest on upper 

slopes of Mt Mabu (TT).


Fig. 16. View over montane forest on Mt Mabu (JB). 

4.3.4 Montane Shrubland (1600‒1700 m) (Figures 18, 19, 20) 

At 1600–1700 m just below the peak there is a limited area of montane shrubland where large 

boulders and rocky slopes are covered by scattered tufts of grass and sedge. Above this the 

summit is exposed and vegetation comprises mostly sedges and shrubby herbs. Such 

vegetation appears to be very typical of exposed granitic peaks across the region. Less time 

was spent exploring this habitat, which covers just a few hectares on the rounded peaks. 

Much of the area is bare rock with patches of small trees and shrubs in sheltered or more 

moisture-rich sites. In these patches Rapanea melanophloeos is the most frequent small tree, 

next to a few stunted Syzygium cordatum, Aphloia theiformis, Maytenus acuminata, 

Aeollanthus buchnerianus, Tetradenia riparia and Dissotis sp. Scattered Aloe arborescens 

were also seen. 

In somewhat more exposed sites, the dominant low shrub, 0.5–2.5 m high is Aeschynomene 

nodulosa, along with Kotschya recurvifolia. Common prostrate or semi-prostrate herbs 

include Ipomoea involucrata, Corrigola drymerioides, Indigofera sp. and Lobelia trullifolia. 

The dominant feature, however, is large clumps of the sedge Coleochloa setifera (Figure 17), 

with smaller clumps of the grasses (?)Danthoniopsis sp. and Helictotrichon elongatum and 

the sedge Cyperus fischerianus. Many of the large clumps had an abundance of the small 

pink-flowered orchid, Polystachya songaniensis (Figure 22).


Fig. 17. Coleochloa 'grassland' 

on rocky slopes near summit of 

Mt Mabu (TT). 

Fig. 18. Upper forest margin with montane 

shrubland and Coleochloa 'grassland' (JT). 

Fig. 19. Montane shrubland near summit of Mt 

Mabu (JT). 

Fig. 20. Montane shrubland near summit of Mt Mabu (JT).


4.3.5 Vegetation of the Drier Western and Northern Slopes 

In the western parts of the massif, which lie in the rain shadow and away from the prevailing 

oceanic moisture-bearing air currents, aerial photographs and study of Google Earth imagery 

suggest that the lower limit of moist forest is around 1200‒1250 m, although extending lower 

to 1050 m on sheltered slopes and along drainage lines and gullies. On the northern boundary 

the lower limit is around 1400 m. Below the forest on this drier side is what appears to be 

woodland and bushland. However, as these areas were not visited it is not possible to confirm 

this. This compares to a lower forest limit of 950 m in gullies and valleys on the southern and 

eastern slopes. 

There appears to be a marked break on the western side of Mabu between higher altitude 

forest (smooth texture, low canopy height) and medium altitude forest (rough texture, varying 

canopy height and colour) at around 1350‒1400 m. This disjunction is not so apparent on the 

moister eastern slopes. 

4.4 Forest Plots 

In order to help characterise forest composition and density, nine 20 × 20 m (0.04 ha) 

enumeration plots were recorded. These were all sited within 1 km of the main campsite, 8 in 

medium-altitude forest and one in adjacent woodland near a small stream. All trees were 

identified to species, and diameter breast height (dbh) for trees above 5 cm dbh recorded 

along with an estimated height. Summarised results are shown in Table 5. 

In the eight forest plots, the main large canopy trees in terms of basal area were (in declining 

order of importance): Newtonia buchananii (primarily in Plot 5), Strombosia scheffleri, 

Chrysophyllum gorongosanum and Maranthes goetzeniana. The main sub-canopy trees (often 

with more stems but lower basal area) were Drypetes gerrardii, Funtumia africana, Rawsonia 

lucida, Drypetes natalensis and a number of Rubiaceae (many being unidentified species but 

also Heinsenia diervilleoides and Aidia micrantha). Plot 1 was situated in the main campsite. 

The total number of stems over 5 cm dbh recorded in the forest plots was 279, giving a mean 

stem density of around 872 stems/ha. Figures for each plot ranged from 50 (= 1250 stems/ha) 

down to 20 (= 500 stems/ha). 

Mean basal area of all trees above 5 cm dbh for the eight forest plots was 29.94 (= 93.56 

m 2 /ha), ranging from 1.97 m 2 (= 49.32 m 2 /ha) to 5.72 m 2 (= 143.0 m 2 /ha). Of this figure, the 

four main canopy trees comprised over 76%, while the main sub-canopy trees (including 

unidentified Rubiaceae) contributed a further 16%, giving a combined total of over 92%. 

In the woodland plot (Plot 9) the major trees were Syzygium cordatum and Parinari excelsa, 

both in terms of number of stems in the plot (11 and 8 respectively) and basal area (equivalent 

to 36.55 and 8.61 m 2 /ha, respectively). Total basal area was 2.0782 (= 51.96 m 2 /ha). There 

were only 4 tree species in the woodland plot in comparison to the forest plots, which had 

from 9 to 17 species (mean 14.5). 

4.5 Observations on Tree Fruiting 

A striking feature of the forest on Mabu during the expedition was the near-absence of figs on 

large stranglers, in contrast to the situation for example in Malawi when October is normally a 

productive month and large hornbills are breeding (Dowsett-Lemaire & Dowsett 2006). Of 

the more than 30 strangler figs examined, only one Ficus bubu was in fruit, one F. scassellatii 

was in unripe fruit and another had ripe figs. Not a single F. thonningii nor F. sansibarica was 

fruiting. The ripe figs of F. scassellatii were taken by Silvery-cheeked Hornbills and Green


Barbets, but observations were of short duration. 

Of other large trees, Olea capensis were generally flowering except for a few trees at the 

forest edge at 1600 m that were out of phase and fruiting (with attendant Rameron Pigeons). 

Cryptocarya were also in flower, and Polyscias were near the end of the fruiting season 

(Rameron Pigeons were seen eating them). Aphloia were in young, fully formed fruit (already 

consumed by Green Barbet and Stripe-cheeked Greenbul). Lower down at 900 m, several 

Xylopia aethiopica were in ripe fruit, the arils being eaten occasionally by Green Barbet, 

White-eared Barbet, Golden-rumped Tinkerbird and Little Greenbul. One pair of Green 

Barbets was breeding there in a territory including many fruiting Xylopia, and the local 

Syzygium cordatum (some flowering) will provide an abundance of fruits in the middle of the 

rainy season. At 1600 m, a tall Syzygium guineense had just finished flowering. The fruiting 

of Pittosporum (common at 1600 m) was finished, with some old capsules still visible on the 

trees. One big Ochna holstii seen at 1600 m had unripe fruit. 

Some Apocynaceae climbers were fruiting (especially Landolphia kirkii, a mammal-eaten 

fruit), and Tabernaemontana stapfiana were also bearing full-grown fruits in several places. 

The liana Schefflera goetzenii had unripe fruit (as on Namuli at this time of year) while Urera 

trinervis had ripe fruit. Stripe-cheeked Greenbuls were also seen taking the ripe, orange 

berries of the climber Rutidea orientalis. Several Garcinia kingaensis were flowering and 

forming young fruit (this species does not flower every year). There were many old fruits of 

Myrianthus holstii on the forest floor, and some "flowers" were seen on the trees (maturation 

is expected in the rains). Some Drypetes gerrardii had fully-formed fruits, not yet ripe. The 

plumed seeds of Funtumia africana were flying around in many places in the forest and open 

pods were seen on trees. Of the smaller understorey species, fruiting is usually spread out, 

with very few ripe fruits at any time – this was true of species of Tricalysia and Coffea, and 

also Synsepalum muelleri and Maytenus acuminata. Others were generally flowering or with 

flower-buds (Canthium sp., Carissa bispinosa, Chassalia parvifolia, Erythroxylum 

emarginatum, Lasianthus kilimandscharicus, Lasiodiscus usambarensis, Pavetta gurueensis, 

Rinorea spp., Tabernaemontana ventricosa), suggesting that the rains will be a productive 

time. Their main dispersers appear to be greenbuls of the genus Andropadus.


Table 5. Summary of forest plot data, Mt Mabu. 

Species Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Plot 7 Plot 8 Plot 9 TOTAL 

no. BA no. BA no. BA no. BA no. BA no. BA no. BA no. BA no. BA stems BA 

Acacia pentagona 1 0.0080 1 0.0080 

Aidia micrantha 4 0.0300 4 0.0300 

Blighia unijugata 2 0.0087 3 0.0548 5 0.0635 

Bridelia micrantha 1 0.0661 1 0.0661 

Chrysophyllum gorungosanum 1 1.9609 1 0.8993 2 1.6178 1 0.2291 1 0.4779 1 0.2291 1 0.0062 8 5.4204 

Cola greenwayi 3 0.0054 10 0.0767 2 0.0048 15 0.0868 

Craibia brevicaudata 1 0.0805 1 0.0805 

Dalbergia boehmii 1 0.0090 1 0.0090 

Diospyros abyssinica 1 0.0022 1 0.0057 1 0.0064 3 0.0142 

Dovyalis macrocalyx 1 0.0275 1 0.0275 

Drypetes gerrardii 1 0.2972 4 0.0486 7 0.8678 5 0.3065 7 0.4990 5 0.0770 7 0.3130 12 0.2570 48 2.6661 

Drypetes natalensis 2 0.0058 2 0.0104 1 0.0106 7 0.0606 1 0.0026 1 0.0064 2 0.0083 16 0.1046 

Englerophytum magalismontanum 3 0.0937 1 0.0133 1 0.0020 5 0.1090 

Erythrococca polyandra 1 0.0045 1 0.0045 

Funtumia africana 1 0.0739 1 0.0087 6 0.0741 1 0.0025 5 0.0700 10 1.0602 11 0.0651 2 0.0099 37 1.3644 

Garcinia kingaensis 1 0.0401 2 0.0581 1 0.0043 4 0.1025 

Haplocoelum foliolosum 1 0.0020 1 0.0020 

Heinsenia diervilleoides 4 0.0380 1 0.0021 1 0.0269 6 0.0670 

Ixora scheffleri 1 0.0050 1 0.0050 

Keetia gueinzii 1 0.0021 1 0.0021 

Lecaniodiscus fraxinifolius 1 0.0025 1 0.0025 

Macaranga capensis 2 0.3878 2 0.3878 

Maranthes goetzeniana 1 0.3167 2 0.0085 2 0.5302 1 1.3072 2 0.1499 1 0.4657 1 0.8172 10 3.5955 

Millettia lasiantha 1 0.0135 3 0.0364 1 0.0021 5 0.0520 

Myrianthus holstii 1 0.0026 1 0.0472 1 0.0145 1 0.0079 3 0.0176 7 0.0898 

Newtonia buchananii 12 4.6231 1 2.6594 13 7.2825 

Oncinotis sp. 1 0.0050 2 0.0117 3 0.0168 

Oxyanthus speciosus 1 0.0054 1 0.0025 2 0.0105 1 0.0036 5 0.0220 

Parinari excelsa 8 0.3444 8 0.3444 

Polysphaeria lanceolata 1 0.0072 1 0.0079 2 0.0151 

Rauvolfia caffra 1 0.0050 1 0.0050 

Rawsonia lucida 1 0.0050 1 0.0028 6 0.2282 1 0.0041 1 0.0021 10 0.2423 

Rubiaceae spp. (uncertain) 1 0.0211 1 0.0284 1 0.0175 4 0.1193 1 0.0064 1 0.0133 1 0.0260 10 0.2320 

Rytigynia uhligii 2 0.0507 2 0.0507


Strombosia scheffleri 2 0.2651 6 4.2483 12 0.7678 3 0.0982 5 0.7412 4 0.3779 32 6.4985 

Synsepalum cf. brevipes 1 0.0106 1 0.0106 

Syzygium cordatum 11 1.4618 11 1.4618 

Tabernaemontana elegans 1 0.0028 3 0.0128 1 0.0661 5 0.0817 

Trilepisium madagascariense 1 0.0106 1 0.0041 2 0.0146 

Vitex doniana 2 0.2631 2 0.2631 

Xylopia aethiopica 1 0.0141 1 0.0141 

unknown 1 0.0804 2 0.6923 1 0.1542 1 0.0255 3 0.1475 8 1.0999 

TOTALS 20 3.0312 33 5.4793 33 3.8886 39 3.1524 50 5.7215 32 1.9728 35 2.2710 37 4.4211 22 2.0782 301 32.0162 

mean BA/ha (m 2 ) equivalent 75.780 136.983 97.215 78.810 143.038 49.320 56.775 110.53 51.955 88.934 

no. = number of individuals greater than 5 cm dbh; BA = basal area (m 2 ) 

Coordinates of forest plots (central point): 

Plot 1 ‒ S16 17 10.3, E36 24 01.2, 981 m 

Plot 2 ‒ S16 16 56.9, E36 23 41.9, 111 m 

Plot 3 ‒ S16 17 00.3, E36 23 47.2, 1097 m 

Plot 4 ‒ S16 17 02.8, E36 23 33.2, 1232 m 

Plot 5 ‒ S16 17 10.1, E36 23 44.8, 1043 m 

Plot 6 ‒ S16 17 14.9, E36 23 55.0, 996 m 

Plot 7 ‒ S16 17 13.3, E36 23 58.8, 998 m 

Plot 8 ‒ S16 17 13.9, E36 23 39.7, 1084 m 

Plot 9 ‒ S16 17 19.6, E36 24 18.6, 956 m


5. PLANTS 


As far as we are aware, no previous biological collecting or survey work (botanical or 

zoological) had been carried out on the Mabu massif prior to our first reconnaissance visit in 

December 2005, nor are any descriptions of the area available. However, specimens cited in 

Flora Zambesiaca indicate that the botanists Barbosa and Carvalho collected in the Tacuane 

area in May 1949, including along the road from Tacuane to Limbuè, and there are also 

numerous collections made from 1943 to 1949 by Helen Faulkner from Namagoa Estates 

(16 o 47' S, 36 o 58' E, alt. 150 m) in Lugela District, 70 km to the south-east. 

Moist forests in Zimbabwe and Malawi have been much better studied than those in 

Mozambique (e.g. Dowsett-Lemaire 1989a, Müller 1999, 2006, White et al. 2001), and 

comparisons can be made to these, particularly to the composition of forests occurring at 

medium altitude (medium altitude and sub-montane according to Müller 1999). However, a 

full comparison across these countries has not yet been done. 

5.2 Plant Collections 

During the Darwin expedition plant specimens were collected from forest, woodland, 

shrubland, rocky outcrops and from the overgrown tea estates across the south-eastern part of 

the mountain. As on previous project expeditions, there was no particular collecting strategy 

other than to gather as comprehensive a collection as possible of fertile identifiable material 

from the full range of accessible habitats. However, particular attention was paid to the main 

forest species. Complete sets were deposited at the National Herbarium in Maputo (LMA) and 

at Kew, while the third and fourth incomplete sets were deposited at the herbarium of the 

Universidade Eduardo Mondlane (LMU) and the National Herbarium in Malawi (MAL). 

There were over 350 numbered collections with notes, with around another 300 voucher 

specimens from survey plots. Nearly all could be identified, and these form the basis of the 

checklist of species found above 800 m given in Annex 3. 

The total number of plant taxa recorded from above 800 m was 249, comprising 9 

Pteridophytes, 1 Gymnosperm, 35 monocotyledons and 204 dicotyledons, covering 90 

families. The largest families in terms of taxa were Rubiaceae (24), Euphorbiaceae (13), 

Leguminosae: Papilionoideae (12), Acanthaceae (12) and Apocynaceae (10). Not 

unexpectedly, the forest understorey families Rubiaceae (mostly in montane forest), 

Acanthaceae and Euphorbiaceae were particularly common. The papilionoids, however, were 

more common in the surrounding woodland. However, the plant collections were quite 

limited, both in extent and intensity. It is probable that the full species list from the Mt Mabu 

area above 800 m altitude will exceed 350 taxa. 

There was much similarity with the flora on Mt Namuli (Timberlake et al. 2009), with 118 

(48%) of the species given here in Annex 3 also being found there. However, the checklist for 

Namuli only covers those species above 1300 m (most of the moist forest on Mt Namuli is 

montane and above 1600 m), whereas most of the forest on Mabu is at medium altitude 

(1000‒1400 m). Medium altitude forest is known to be more species-rich than montane forest 

(T. Müller, pers. comm.).


5.3 Species of Particular Interest 

The species of particular interest comprise those that are new, endemic, threatened or of 

particular conservation concern, and those that are new records or significant range 

extensions. There were 18 species of particular interest recorded from Mt Mabu during the 

expedition. These are shown in Table 6. 

Table 6. Plant species of interest recorded from Mt Mabu. 

Family Species notes 

Amaryllidaceae Cryptostephanus vansonii previously a E Highlands/Moz border endemic 

Orchidaceae Bulbophyllum ballii 1st record for Moz; previously a Zimbabwe 

endemic 

Orchidaceae Bulbophyllum sandersonii 1st record for Moz 

Orchidaceae Polystachya malilaensis 1st record for Moz 

Orchidaceae Polystachya songaniensis previously thought endemic to Mts Mulanje & 

Zomba; 1st record Moz 

Poaceae Oreobambos buchwaldii 1st record for Moz 

Xanthorrhoeaceae Dianella ensifolia 

in FZ area previously only known from 

Chimanimani Mts 

Acanthaceae Mimulopsis arborescens 1st record for FZ area; signficant range 

extension from S Tanzania 

Acanthaceae Justicia asystasioides significant range extension from N Malawi 

Acanthaceae Sclerochiton hirstus only 2nd collection, previously thought to be a 

Namuli endemic 

Asteraceae Bothriocline glomerata 2nd FZ record (+ Namuli) 

Euphorbiaceae Crotonogynopsis usambarica new genus for FZ area (previously Tanzania) 

Loranthaceae Helixanthera schizocalyx new species; Mabu endemic 

Molluginaceae Corrigiola drymerioides 2nd record for Moz (Namuli) 

Rubiaceae Didymosalpinx norae 2nd record for Moz (Garuso) 

Rubiaceae Rytigynia sp. not matched at K 

Rutaceae Vepris sp.nov. near V. bachmannii possible new sp. 

Viscaceae Viscum cylindricum 1st record for Moz (previously Mal + Tanz) 

Fig. 21 a,b. Cryptostephanus vansonii on upper slopes 

of Mt Mabu (JT). 

Fig. 22. Polystachya songensis 

orchid on Mabu summit (TT).


New Species and Records 

Of these 18 listed species, two have just been described (Helixanthera schizocalyx, Harris et 

al. 2011) or are thought to be new (Vepris sp. nov.), while five represent significant range 

extensions (Cryptostephanus vansonii (Figure 21) and Dianella ensifolia from the 

Chimanimani Mountains 600 km to the south; Mimulopsis arborescens, Justicia asystasioides 

and Crotonogynopsis usambarica from N Malawi or S Tanzania 700 km to the north). There 

are twelve species that are new records for Mozambique, of which four are known from Mt 

Mulanje in Malawi (Strugnell 2006), so their occurrence in a similar habitat nearby is not 

surprising. 

Although its presence is not surprising, and probably more a reflection of under-recording in 

Mozambique rather than a particularly signficant range extension, it is interesting to note that 

this is the first record of the large common forest bamboo, Oreobambos buchwaldii, for 

Mozambique. 

Didymosalpinx norae was previously only known from Chirinda Forest in SE Zimbabwe 

(Timberlake & Shaw 1994) and from Garuso Forest near Chimoio in central Mozambique, as 

well as in a few lowland forests in Kenya and Tanzania. It discovery almost 400 km northeast 

from its previous known southern African populations is quite unusual for a showy 

flowering shrub. 

A question often raised is whether the Mulanje Cedar, Widdringtonia whytei, is found on 

similar mountains in Mozambique; to date it has only been found in Malawi. There was no 

sign of either W. whytei or, surprisingly, the more widespread W. nodiflora on Mabu, or on 

any other Mozambican mountains that have been visited during the Darwin project, although 

the massif may be too low and warm for it. However, it has been recorded from Mt 

Gorongosa in the centre of the country at around 1500 m (Müller et al. 2008). 

It was interesting to note the abundance of Maranthes goetzeniana, a large forest tree that to 

date does not appear to have been recorded from any forest in Malawi (White et al. 2001). It 

was also not found on Mt Namuli (Timberlake et al. 2009), although it is common in midaltitude 

forests in central Mozambique and adjacent parts of Zimbabwe. 

Threatened and Endemic species 

Apart from the newly-described mistletoe and possible new Vepris, it appears that there are 

no other endemic plant species on Mt Mabu, nor any taxa that are known to be particularly 

threatened across their range. This is in marked contrast to the situation on Mt Namuli where 

16 endemic plant taxa, including five new species, were recorded (Timberlake et al. 2009) 

and, in addition, 26 new Mozambique records. 

From the Sabonet Red Data List (Golding 2002, Izidine & Bandeira 2002, Timberlake et al. 

2006) there are 14 species on the Mabu checklist that are said to be endemic to the Flora 

Zambesiaca area, Mozambique or adjacent countries, or were on one or more country's Red 

Lists (Table 7). However, a number of these have since been shown to be not as restricted in 

distribution as was originally believed or were cited in error. 

What is apparent is that, compared to the zoological findings, the flora of Mabu has few 

species of particular interest; most species found here are moderately widespread in the 

scattered patches of most forest found across Eastern and Southern Africa. What is of much 

greater botanical interest, however, is the extent and good condition of the moist forest at an 

altitude where, elsewhere, so much of it has been cleared. For example, the total extent of 

moist forest on Mt Mabu is around 7880 ha, of which 5270 ha lies between 1000 and 1400 m, 

compared to 1300 ha of similar forest on Mt Chiperone (Timberlake et al. 2007), 135 ha on


Mt Namuli (Timberlake et al. 2009) and an unknown extent (perhaps 1000‒2000 ha, figure 

not given in Müller et al. 2008) on Mt Gorongosa. 

Further survey work should focus on the medium altitude forest areas, particularly gullies and 

streams within the forest and on the gully or riverine forests at lower altitudes. The drier side 

of the mountain has not yet been investigated biologically, and should contain some different 

habitats and species from the moister eastern side. 

Table 7. Species found on Mt Mabu that are mentioned on the Sabonet Red Data lists (Izidine & 

Bandeira 2002, Golding 2002) as threatened or endemic. 

Family Species RD or endemic status Notes 

Aloaceae Aloe arborescens VU in Mw common elsewhere 

Amaryllidaceae Cryptostephanus vansonii FZ endemic 

Orchidaceae Bulbophyllum ballii Zw endemic; VU in Zw now found in Moz 

Orchidaceae Polystachya songaniensis Mw endemic, LR now found in Moz 

Acanthaceae Brachystephanus africanus CR in Zw more common elsewhere 

Acanthaceae Sclerochiton hirsutus Moz endemic, DD 

Asteraceae Senecio peltophorus Mw endemic; VU now found in Moz 

Moraceae Ficus bubu CR in Zw more common elsewhere 

Moraceae Ficus sansibarica VU in Zw more common elsewhere 

Moraceae Ficus vallis-choudae EN in Zw more common elsewhere 

Rubiaceae Coffea mufindiensis subsp. FZ montane endemic common 

australis 

Rubiaceae Tricalysia acocantheroides VU in Zw 

Rutaceae Zanthoxylum gilletii VU in Zw more common elsewhere 

Sapindaceae Allophylus chaunostachys VU in Zw


6. BIRDS 


Prior to the survey carried out in October 2008, the birds of Mabu were known only through a 

brief visit by Claire Spottiswoode and Eric Herrmann on 10–14 December 2005 

(Spottiswoode et al. 2008), during which some 50 bird species were observed in the lowaltitude 

tea forest as well as in the main forest block up to 1220 m. There are no other bird 

records known from the area. Jack Vincent in his trip to the nearby Mt Namuli (Vincent 

1933b, 1933‒1936) did not pass close to Mabu. 

6.2 Bird Survey 

During the present survey, from 10–30 October 2008, some 120 bird species were recorded, 

bringing the total bird list to 126 species (Annex 4). Four nights were spent at 550 m altitude 

by the old tea estate manager's house (10–12 & 29 October) when riparian forest, mixed tea 

forest and other habitats in the vicinity were explored, and 16 nights in the forest, mainly 

around 1000 m (main forest camp), with four nights also at higher levels (near the summit at 

1400 m and north-west of the main camp at 1300 m). 

In addition, some mist-netting was carried out over two days in order to assess state of moult 

and breeding condition (see Annex 5). All birds were released after being marked with rings 

from the East African scheme (Kenya National Museum). 

The sections below discuss the biogeographical and conservation significance of the forest as 

related to birds (see also Dowsett-Lemaire 2010 for a comparison of Mts Mabu and Namuli). 

Nomenclature follows Dowsett-Lemaire & Dowsett (2006). A more comprehensive account 

of Mabu's birds is given in Dowsett-Lemaire & Dowsett (2009). 

6.3 Biogeographical Significance 

The known distributions of Afromontane and selected Eastern bird species on the five main 

montane massifs of the northern Mozambique/southern Malawi region are shown in Table 8. 

These are, from north to south, Namuli, Mulanje, Thyolo, Mabu and Chiperone. Records from 

Thyolo in southern Malawi are largely historical, as the forest was almost totally destroyed in 

the early 2000s. Records from Chiperone are based on Benson (1950) and Spottiswoode et al. 

(2008). The mention of Cabanis's Greenbul (race placidus) comes from a specimen duly 

collected by J. Makawa (R. Dean & R. Prys-Jones, pers. comm. 2009) but missed from the 

1950 publication. Chorological status follows Dowsett-Lemaire & Dowsett (2006). 

The biggest and tallest massif, Mulanje, has the highest number of Afromontane species, 31 

(including the Spotted Ground Thrush), with Namuli coming a close second (27). A few nonforest 

species are not recorded (nor expected) to occur on Mabu, such as the Red-tailed 

Flufftail Sarothrura affinis, Blue Swallow Hirundo atrocaerulea and Cinnamon Bracken 

Warbler Bradypterus cinnamomeus. These absences can be explained by the reduced size of 

grassland or shrubland habitats. Although the Namuli massif has more extensive open 

habitats, these same birds are also missing there (the flufftail almost certainly so, Dowsett- 

Lemaire 2008) ‒ the dominant type of grassland on Namuli gets water-logged and there are 

apparently no suitable breeding banks for Blue Swallow, while the montane shrubland is 

floristically impoverished and of limited extent for a species such as the Bracken Warbler. 

The total number of Afromontane birds recorded on Mabu is 18 species. A few high-altitude 

montane forest birds seem to be absent (e.g. Olive-flanked Robin Cossypha anomala, Whitetailed 

Crested Flycatcher Elminia albonotata, Eastern Double-collared Sunbird Nectarinia 

mediocris), all of which are recorded from the higher-altitude forests of Mulanje, Namuli and


Table 8. Afromontane (near-)endemic and selected Eastern endemic bird species present on Namuli, 

Mulanje, Thyolo, Mabu and Chiperone Mountains. Species are all found in forest except those marked 

"NF" (not forest) where the habitat consists of grassland, rocks, montane shrubland or forest edges. 

Afromontane species: Namuli Mulanje Thyolo Mabu Chiperone 

Red-tailed Flufftail (NF) ? X - - - 

Rameron Pigeon X X X X X 

Cinnamon Dove X X X X X 

Cape Eagle Owl (NF) X X - - - 

Scarce Swift X X X - - 

Bar-tailed Trogon X X X X X 

Moustached Green Tinkerbird - X - - ? 

Blue Swallow (NF) - X - - - 

Grey Cuckoo-shrike - - X X X 

Eastern Mountain Greenbul X X - - - 

Stripe-cheeked Greenbul X X X X X 

Cabanis's Greenbul X X X X X 

Olive Thrush X X X - X 

Orange Thrush X X X - X 

Cholo Alethe X X X X X 

Starred Robin X X X X X 

Swynnerton's Robin - - - X - 

Olive-flanked Robin X X - - X 

Cape Robin (NF) X X - X - 

Cinnamon Bracken Warbler (NF) - X - - - 

Evergreen Forest Warbler X X X - X 

Yellow-throated Warbler X X X X X 

Wailing Cisticola (NF) X X - X - 

Bar-throated Apalis - X - - - 

Namuli Apalis X - - X - 

Cape (Malawi) Batis X X X X X 

White-tailed Crested Flycatcher X X X - X 

Dapple-throat X - - X - 

Eastern Double-collared Sunbird X X - - X 

Olive Bush Shrike - X - - - 

Bertram's Weaver (NF) X X X X X 

Red-faced Crimsonwing X X X - X 

Swee Waxbill (NF) X X X X X 

African Citril (NF) X X X - X 

Sub-Afromontane species: 

Spotted Ground Thrush X X X X - 

Selected Eastern Species: 

S. Banded Snake Eagle - - - X - 

White-eared Barbet X X X X X 

Green Barbet X - X X - 

Eastern Green Tinkerbird X - - - ? 

Gunning's Akalat - - - X - 

White-winged Apalis X X X - X 

Green-headed Oriole - - X X X


Chiperone. The absence of the Evergreen Forest Warbler Bradypterus lopezi is more 

surprising as it is common in mid-altitude forest throughout SE Malawi and on Chiperone 

where it reaches its southern limit of range (Dowsett-Lemaire 1989b), but elsewhere in 

Malawi it can be rare in this forest type (e.g. Ntchisi Mountain). Even more surprising is the 

apparent absence of Orange Thrush Zoothera gurneyi as this species is also common in midaltitude 

forests in the region. It is unlikely to have been overlooked as it is very vocal from the 

months of at least September, even in dry weather (and it is unknown to the local hunters, see 

species account of Zoothera guttata). It is possible that further exploration of the high-altitude 

sections of Mabu forest (the largest area of which was not visited) will reveal the presence of 

some additional discreet Afromontane species such as Red-faced Crimsonwing Cryptospiza 

reichenovii. The intra-African migrant Scarce Swift Schoutedenapus myoptilus might have 

been missed as this noisy bird normally arrives by October, but could have been delayed by 

drought. 

Many of the Afromontane birds listed above reach their overall southern limits of range 

within this region: seven (Bar-tailed Trogon Apaloderma vittatum, Cabanis's Greenbul, Oliveflanked 

Robin, Evergreen Forest Warbler, Eastern Double-collared Sunbird, Bertram's 

Weaver and African Citril Serinus citrinelloides) reach the most southerly massif of 

Chiperone. Mulanje represents the southern limit for Moustached Green Tinkerbird, Eastern 

Mountain Greenbul and Cinnamon Bracken Warbler, while Dapple-throat, which skips 

Malawi altogether, reaches Mabu, an extension from Namuli. 

A small cluster of species (Afromontane or Eastern or otherwise) are shared between Thyolo, 

Mabu and Chiperone: Bronze-naped Pigeon, Grey Cuckoo-shrike and Green-headed Oriole. 

They all reappear on Mt Gorongosa, some 300–350 km to the south-west (Oatley & Tinley 

1989). Mt Chiperone remains under-explored, especially the higher levels not reached by 

Spottiswoode et al. (2008). In addition to identifying the green tinkerbird, a form of Barthroated 

Apalis (the yellow-bellied race of Mulanje or Namuli Apalis) should be sought in the 

montane forest. 

The presence of Gunning's Akalat on Mabu suggests that the local microclimate is fairly 

warm, also shown by the distribution of some lowland forest species there ‒ White-throated 

Nicator, Red-capped Robin and Blue-mantled Flycatcher all reach an altitude of 1400 m, 

whereas on the wet southern slopes of Mulanje they do not ascend above 800‒950 m. On 

Namuli the Nicator does not ascend above 1160 m (J. Graham, pers. comm.), the Red-capped 

Robin reaches 1200 m and the flycatcher has not been recorded (if it occurs, this would be 

below 1150 m). Similarly the warm microclimate of Mabu could explain the scarcity of 

montane species such as Namuli Apalis and Dapple-throat, and the absence of high-montane 

species such as Olive-flanked Robin (very common on the large, colder massifs of Mulanje 

and Namuli above 1000 and 1550 m respectively). 

6.4 Bird Species of Conservation Importance 

The forests on Mt Mabu are especially important for the following Red Listed species 

(BirdLife International 2008) and the belcheri population of Green Barbet: 

Southern Banded Snake Eagle Circaetus fasciolatus. Near Threatened. Several territorial 

pairs inhabit lower levels of the forest, including the foothills. It is an Eastern endemic, 

found mainly in coastal forest and unrecorded from adjacent Malawi, hence its discovery 

at Mabu represents a small range extension. 

Green Barbet Stactolaema olivacea belcheri. An Eastern endemic with a very patchy 

distribution on the eastern side of Africa from coastal Kenya to Natal. This well-marked


race was until recently thought to be confined to the forests on Thyolo and Mt Namuli. 

Forests on Thyolo Mountain have been eliminated by recent deforestation (Dowsett- 

Lemaire & Dowsett 2006) and the barbet numbers on Namuli are rather small, at most 

30–40 pairs (Dowsett-Lemaire 2008). First discovered on Mabu by C. Spottiswoode in 

2005, this barbet occurs over the full altitudinal range of forest, and locally down to 750 

m. Thus Mabu harbours the most important population to date, of the order of several 

hundred pairs (probably over 500 pairs; individual territories often cover about 10 ha). 

Spotted Ground Thrush Zoothera guttata. Endangered. A rare thrush of mid-altitude forest in 

tropical Africa (i.e. a sub-Afromontane endemic), common only in the temperate forests 

of eastern South Africa (Chittenden in Hockey et al. 2005). It is known to have 

movements down to the coast in the off-season (although the populations in S Malawi 

appear to be resident). Until recently in Mozambique it was known only from a couple of 

winter records on the coast (Parker 2005), but was discovered in the breeding season on 

Namuli (Dowsett-Lemaire 2008). Its numbers on Mabu are probably very low, but the 

near-absence of song in October means that proper evaluation of densities will have to 

await further study. 

Cholo Alethe Alethe choloensis. Endangered. Confined to mid-altitude and lower 

Afromontane forests in SE Malawi and adjacent N Mozambique. Populations in Malawi 

have decreased due to deforestation (especially bad on Thyolo Mountain, the lower slopes 

of Mulanje and at Lisau and Soche); they have also decreased on Namuli through the 

near-eradication of mid-altitude forest. Figures proposed for Namuli by Ryan et al. (1999) 

are far too high as the bird is rare in the cooler forest of Manho, the largest block of forest 

on Namuli (Dowsett-Lemaire 2008). The species occurs over a wide altitudinal range on 

Mabu (950–1650 m) but is rather uncommon below 1200 m. Vocal activity in most of 

October was low owing to the drought and population estimates are difficult to propose, 

although the order of magnitude should be around a 1000–1500 pairs. It is possible there 

are more Alethes on Mabu than on the whole of Mulanje (where the area of suitable forest 

covers c.5000 ha). In any case Mabu can be considered as extremely important for this 

species. 

Swynnerton's Robin Swynnertonia swynnertoni. Vulnerable. This Afromontane endemic was 

known only from the forests of eastern Zimbabwe and adjacent Mozambique (including 

Gorongosa) and the central highlands of Tanzania (mainly Udzungwa, with few in the 

East Usambaras). Its discovery on Mabu partly fills the gap between the Zimbabwe/S 

Mozambique and Tanzanian populations as Mabu is about 350 km NE of Gorongosa. On 

Mabu it is confined to the higher-altitude forest above 1340–1400 m and is not uniformly 

distributed as it favours dense understorey. The area of forest above the contour line of 

1400 m is about 900 ha, so the total population may be of the order of 100‒200 pairs. 

Gunning's Akalat Sheppardia gunningi. Near Threatened. An Eastern endemic with a patchy 

distribution from coastal Kenya to coastal Mozambique, and an inland population on the 

northern lakeshore of Lake Malawi and the eastern scarp of the Viphya Plateau. First 

discovered on Mabu by C. Spottiswoode in 2005, it is now seen to occur over a wide 

altitudinal range from 400–1350 m. It is surprising that this species has adapted to pure 

tea understorey under forest canopy (as has another understorey species, the Blue-mantled 

Flycatcher Trochocercus cyanomelas), and a substantial part of the population will be 

destroyed if the tea farms are reopened. Large numbers will, however, survive in the 

natural forest at low and middle levels if it receives adequate protection. The patchy 

habitat makes density estimates difficult, but there are probably several hundred pairs. In 

May 2009 L. Fishpool discovered the species on Mt Inago in forest patches at 1050‒1450 

m (Fishpool & Bayliss 2010). Inago (peak 1961 m) is 45 km to the north-east of Namuli


and 180 km NE of Mabu. The fairly richly coloured underparts and measurements of the 

Mabu birds (wing-lengths 71.7 mm, n=4, Appendix 5, compared to 71.3 mm, range 67.5– 

74.5 mm, n=9, from Malawi, R.J. Dowsett, pers. comm.) suggest that these birds are more 

closely related to the race bensoni of the northern Malawi lakeshore than to coastal races, 

despite the fact that the nearest population to Mabu is situated on the Mozambique coast 

at c.18°S. 

Namuli Apalis Apalis (thoracica) lynesi. Near Threatened. Until now this form of Barthroated 

Apalis was considered endemic to the forests of Mt Namuli, being replaced in SE 

Malawi by the yellow-bellied form flavigularis. Unlike on Namuli where it is very 

common, it is distinctly rare on Mabu, found only above 1400 m. Two of the three males 

heard or seen were unmated, which suggests that the population is not thriving. Possibly a 

warm microclimate combined with a very tall forest canopy are not favourable for this 

montane species, which prefers short-canopy forest and rich shrubland. The population is 

certainly small (a few dozen pairs, up to 100?). But the occurrence of this taxon on Mabu 

means that other high-altitude forests in this part of Mozambique are likely to harbour it, 

including a couple of high peaks next to Namuli and the southern section of the Namuli 

plateau (south of the Rio Malema dambo, with many forest patches around 1500 m), and 

possibly even Mt Chiperone, the upper parts of which remain largely unexplored. It was 

not, however, discovered on Mt Inago by L. Fishpool in 2009, where the altitude of the 

main forested area is rather low (below 1500 m). 

Dapple-throat Modulatrix orostruthus. Vulnerable. This Afromontane endemic was thought 

to reach its southern limit of distribution from the highlands of central Tanzania on Mt 

Namuli, but its presence on Mabu extends its range by some 160 km to the south-west. 

Although quite common on Namuli, on Mabu it is rare and highly localized and limited to 

the Afromontane forest above 1400 m. The species normally favours understorey with a 

dense layer of saplings under closed canopy; the disturbed canopy in parts of the forest, 

caused partly by temporary habitations during the civil war, may in part explain its rarity. 

Only three territorial birds were located (as with the Namuli Apalis), one of which did not 

even react to tape playback. One bird (paired) followed for 3 hours covered an area of at 

least 5–6 ha. Overall numbers must be low, of a few dozen pairs or up to 100. 

6.5 Bird Breeding Records 

October is normally an important month for laying in forest birds in SE Africa (Dowsett & 

Dowsett-Lemaire 1984, Dowsett-Lemaire & Dowsett 2006), but that does not appear to be the 

case on Mt Mabu, at least in 2008. Clearly breeding had started in some species, but some 

pairs were perhaps delayed because of the prolonged drought. Species foraging on the ground 

such as alethes and akalats normally breed as soon as the main rains start and driver ant 

foraging activity increases; the few mist-netted in mid-October (Annex 4) were still inactive. 

Cinnamon Dove: one bird was already sitting on completed nest, despite absence of eggs (27 

Oct), thus egg-laying probably end Oct. 

Green Barbet: one pair incubating (until at least 21 Oct), then feeding (by 28 Oct) = egglaying 

in Oct. 

Starred Robin: judging by the behaviour of pairs or males, breeding had just started in several 

pairs (not yet feeding, incubation stage) = egg-laying in Oct in several cases. 

Black-headed Apalis: pair feeding fledgling(s) 26 Oct (= laying Sept). 

Mozambique Batis: male feeding incubating female, 30 Oct (500 m) = laying Oct. 

Olive Sunbird: one nest-building in the canopy 21 Oct [L. Fishpool] 

Southern Puffback: nest-building 30 Oct (= laying Nov).


Mist-netted Green Twinspots had completed moult (one was just finishing), which suggests 

breeding at the end of the rains (normal in a seed-eater). Outside forest (transition 

woodland), a Chestnut-bellied Kingfisher was incubating 29 Oct (= laying Oct). 

Fig. 23. Olive Sunbird in Mt Mabu forest (JB).


7. OTHER VERTEBRATES & INVERTEBRATES 


Prior to the survey carried out in October 2008 there had been limited recording of mammals 

and herps, apart from some opportunistic observations by Julian Bayliss on reconnaissance 

trips (late 2005/early 2006, mid 2008), although the butterflies were looked at in greater detail 

on these trips. During the October 2008 expedition, detailed studies were carried out on the 

avifauna (see Section 6) to complement earlier work in 2005 (Spottiswoode et al. 2008) and 

systematic survey work was done on the butterflies of the forest and summit areas as well as 

detailed collecting of reptiles and bats. Collecting was more opportunistic for amphibians, 

other small mammals, freshwater crabs and molluscs. All findings are given below by group. 

7.2 Larger Mammals 

The main source of information on larger mammals was interviews by R.J. Dowsett in 

October 2008 with a local hunter, Mr Ofelio Kavaliyawo using an illustrated field guide (R.J. 

Dowsett in Dowsett-Lemaire & Dowsett 2009). The hunter appeared knowledgeable and had 

lived in the area for a number of years. He regularly set traps in the area surrounding the main 

forest camp and was employed as a guide in late 2005 and all subsequent visits. In addition 

any sightings or aural records were recorded. Full details are given in Annex 5. 

The Blue Monkey Cercopithecus albogularis is common within the forest but is hunted by the 

local community using bow and arrows, while Grant's Bush Baby Galagoides (zanzibaricus) 

granti was heard calling through the forest at night, especially around the main forest camp. 

Forest antelopes are under severe threat from bush meat hunting, primarily using gin-traps. 

Species such as Blue Duiker Cephalophus monticola and Bushbuck Tragelaphus scriptus are 

particularly hunted, as well as Klipspringer Oreotragus oreotragus and species such as 

Hyraxes (Procavia capensis and Heterohyrax brucei). According to local hunters, Leopard 

Panthera pardus are occasionally encountered. Both Buffalo Syncerus caffer and Elephant 

Loxodonta africana were historically common, although elephant have not been seen in 

recent years and the last sighting of buffalo was of a large herd passing through the tea estate 

in 2002. 

7.3 Bats 

As far as we are aware, there are no previous bat records from Mt Mabu or the immediate 

area. The nearest well-sampled and comparable localities in terms of habitat and environment 

are the montane areas of southern Malawi (Happold & Happold 1987, 1997) and Mt Namuli 

(see Timberlake et al. 2008, Curran & Kopp 2009). Although bats were opportunistically 

collected earlier by J. Bayliss, a more systematic survey was undertaken by Michael Curran 

and Mirjam Kopp in October 2008 (Curran & Kopp 2009). There have been some additional 

collections since (J. Bayliss), and the results presented here reflect all these findings. Records 

are backed up by voucher specimens deposited in three collections in the Durban Natural 

Science Museum (DNSM), the Museum of Natural History Geneva (MNHG) and the Natural 

History Museum Maputo (NHMM). Nomenclature follows Monadjem et al. (2010b). 

Methods 

Bats were sampled at four sites (Table 9) using mist nets, harp traps and acoustic monitoring. 

At each sampling site two to four ground nets (6 × 3 m and 9 × 3 m) and a canopy fixture 

were employed, the latter consisting of four 9 m nets (monofilament nylon with a 70/2 denier) 

hoisted to 16 m at site 3 and two 6 m nets held at 7.5 m height at site 4. Ground and canopy 

nets were opened from sunset until midnight or later, depending on the level of bat activity. 

Harp traps (2-tier, 1.5 × 1.5 m) were deployed from sunset to sunrise (18:00–05:00). Acoustic


monitoring was carried out at throughout the night consisting of one peak time (c.18:00) 

transect of 100 m lasting 10–15 minutes followed by two additional transects between 20:00 

and 22:00. Bat calls were recorded from a D240 detector recording a sampling frequency of 

44.1 kHz. Hand-held reference calls (Rhinolophidae and Hipposideridae) and hand-released 

calls (all other families) were taken from each species. 

Table 9. Location of bat sampling sites and sampling effort. 

Sampling site alt. (m) long. (S) lat. (E) no.nights 

1. Base camp 550 16.3058 36.4242 2 

2. Main forest camp 980 16.2864 36.4030 4 

3. River near main camp 1000 16.2848 36.3980 5 

4. Forest satellite camp 1300 16.2775 36.3803 2 

Diversity and assemblage structure 

Twelve species of bats across 38 individuals were recorded over the October 2008 sampling 

period (Annex 7). The mid-altitude forest assemblage superficially resembles that of the 

nearby montane forests of Mt Mulanje (Curran & Kopp, unpublished data) in containing a 

large proportion of species from the guild of 'dense-clutter feeders' (Kalko 1998), including a 

dominance of species from the families Rhinolophidae and Hipposideridae. Also of note is the 

presence of one member of the genus Kerivoula (woolly bats), a group which is generally 

viewed as being forest specialists. The identity of this particular Kerivoula species is 

uncertain at this stage, although based on external and cranial measurements it is clearly 

distinct from the two recognized Southern African species, K. argentata and K. lanosa. It 

represents an important record for Mozambique, either acting as the most southerly range 

extension for an existing species or possibly a new species in its own right, and indicating that 

montane forests are forming important stepping-stones for forest-restricted bat species. A new 

species of horseshoe bat, Rhinolophus mabuensis (Taylor et al. 2012, previously lumped with 

R. hildebrantii), was collected from Mabu and also later from neighbouring Mt Inago (Bayliss 

et al. 2010, Monadjem et al. 2010a). 

In terms of abundance and dominance, activity levels (as measured in mist net captures) were 

unexpectedly low over the 13 nights of sampling in comparison to similar altitudes and 

habitats on Mts Mulanje and Namuli. This low abundance was confirmed by acoustic data 

from echolocation transects. It could be due to seasonal aspects such as low levels of insect 

activity, or a low proportion of fruiting trees in the case of fruit bats. A species accumulation 

curve of the current data shows incomplete sampling suggesting that the low abundance is not 

indicative of low species richness or diversity and that many more species have yet to be 

discovered. 

7.4 Other Small Mammals 

Opportunistic small mammal recording was mostly undertaken by Julian Bayliss and Lucas 

Sabão. Bucket pitfall traps were positioned in different habitat types ranging from moist forest 

to the forest edge. A total of 17 specimens, consisting of 4 species of rodents and 3 species of 

shrew, were collected from 1000 to 1300 m in October 2008 and June 2009 (Annex 7). 

Specimens were sent to Peter Taylor at the Durban Natural Science Museum; nomenclature 

follows Musser & Carleton (2005). 

Specimens were prepared as skins and skulls in the collection of the Durban Natural Science 

Museum. Identification was based on keys in Skinner & Chimimba (2005) and the online key


of the Mammals of Tanzania (Field Museum of Natural History: 

http://www.fmnh.org/tanzania/). 

All four rodents are tropical forest specialists and are affiliated with montane or coastal 

forests of central (Malawi) and eastern Africa (Kenya, Uganda, Tanzania, DRC). The 

populations sampled (except for Grammomys dolichurus) are the most southerly known. The 

dominant rodent was the soft-furred mouse Praomys delectorum, a montane forest specialist. 

Although in the IUCN 2010 Red Data List Beamys major (Figure 24) has been included in the 

widespread B. hindei, and Lophuromys aquilus in the widespread L. flavopunctatus, leading to 

them being placed in the category of Least Concern, it is highly probably that the southern 

populations are distinct species (Musser & Carleton 2005) meaning that they may qualify for 

Data Deficient or Threatened status. To the best of our knowledge B. major has not been 

recorded previously for Mozambique, although it was suspected to occur by Smithers & Tello 

(1976). 

The shrew fauna of Mt Mabu is affiliated with tropical forests and woodlands as both 

Crocidura luna and C. olivieri found on Mt Mabu reach their southern extent in Africa in the 

montane forests of the Eastern Highlands of Zimbabwe. Both species are widespread in 

tropical Africa and are listed as Least Concern in the 2010 IUCN Red List. 

Similar studies conducted in nearby montane forests (Mts Chiperone, Namuli and Mulanje) 

suggest that the small mammal fauna of Mt Mabu has been under-sampled at this stage. 

Comparison with Durban Museum collections from additional montane forest habitats 

between 2006 and 2009 in northern Mozambique (Mts Chiperone, Namuli and Inago) indicate 

that six additional species of rodents should also be found on Mabu: Acomys spinossisimus, 

Aethomys namaquensis, Dendromus melanotis, Dasymys incomtus, Mus minutoides and 

Otomys angoniensis. Whilst P. delectorum, the dominant rodent on Mt Mabu, was also found 

on Chiperone and Namuli, the other Mabu forest-specialists (G. dolichurus, B. major and L. 

aquilus) were not, although this may be due to under-sampling. Amongst the shrews, C. 

silacea and C. luna were also found on Mt Namuli but an additional species (C. mariquensis) 

collected on Namuli can be expected to occur on Mabu. 

As with the bats, the terrestrial small mammal fauna occurring in these isolated relic montane 

forests of northern Mozambique form an important southern refuge biogeographically linked 

to tropical montane and coastal forests of central and east Africa. 

Fig. 24. Lesser Pouched rat (Beamys major) caught in medium altitude forest, Mt Mabu (JB). 

7.5 Herpetofauna 

The herpetofauna were opportunistically sampled by Julian Bayliss between 2005 and 2008, 

resulting in the discovery of several new species, and more systematically by Bill Branch and 

Werner Conradie in May 2009 (Branch 2011). To date, 7 amphibian species and 15 reptile


species (9 lizards, 6 snakes) have been recorded (Annex 8). The relatively low species count 

is a result of low sampling effort with surveys occurring during periods of reduced activity. 

Much greater herpetofaunal diversity can be expected with further study. 

Three new species of reptile have been discovered including a forest viper (Atheris 

mabuensis, Figure 25) in 2006 (Branch & Bayliss 2009), the southernmost record of the 

genus; a chameleon, Nadzikambia baylissi, discovered in 2008 and belonging to a genus 

previously thought to be endemic to Mt Mulanje (Branch & Tolley 2010); and also in 2006 a 

pygmy chameleon (Rhampholeon sp., Figure 26) that awaits description (Branch et al. in 

prep.). In addition to these new species, the taxonomic status of some of the other species is 

being investigated, including the status of a rare burrowing skink (Melanoseps sp.), again at 

its southernmost limit, an unusual large-scaled green snake (Philothamnus cf. carinatus), and 

two cryptic leaf-litter frogs (Arthroleptis sp.). Also of note is an unusual olive night snake 

(Dipsadoboa) that managed to escape after capture and which may prove to be new. 

A number of species are shared with Mt Mulanje in southern Malawi, including the frogs 

Leptopelis flavomaculatus and Hyperolius puncticulatus, the chameleon Triceros melleri, and 

the snakes Natricteres sylvatica and Naja melanoleuca. Closely-related chameleons occur on 

both mountains with Nadzikambia mlanjensis and Rhampholeon playceps on Mt Mulanje and 

the sister taxa N. baylissi and R. sp. nov. on Mt Mabu. No forest viper (Atheris) has been 

recorded from Mt Mulanje, whilst A. mabuensis is now known from both Mabu and Namuli. 

Numerous Mt Mulanje endemics, such as the lizards Trachylepis mlanjensis, Proscelotes 

mlanjensis, Platysaurus michelli, Lygodactylus rex and L. bonsi, and the frogs Arthroleptis 

francei, Amietia johnstoni, Ptychadena broadleyi, Strongylopus fuelleborni and Notophryne 

broadleyi, have yet to be found on Mabu, although some are known to occur on Mt Namuli 

(e.g. Lygodactylus rex and Strongylopus fuelleborni). 

Fig. 25. Two Atheris mabuensis on forest floor (JB). 

The Gaboon Viper (Bitis gabonica) was collected from a relatively low altitude in the 

overgrown tea forest but is suspected to also occur in the main forest area. It was previously 

recorded from Mt Chiperone as part of an earlier survey (Timberlake et al. 2007), the first 

record since 1950 for this region of Mozambique. The Gaboon Viper is not known from


southern Malawi, but has been recorded from the Mzuzu‒Nkhata Bay area in the north 

(Broadley pers. comm.). 

The results generally indicate that many of the species are at their southernmost range with 

affinities to groups from the north and west. 

Fig. 26. Pygmy chameleon, sp. nov. (JB). 

7.6 Lepidoptera 

The butterflies of Mt Mabu have been studied on a number of occasions, notably December 

2005, January 2006, June 2008, September 2008, October 2008 and November 2010. 

Specimens were collected by J. Bayliss, C. Congdon, S. Collins, M. Hassan, I. Bampton, R.J. 

Dowsett and S. Georgiadis from a range of habitats including the abandoned tea estate, 

woodlands at the base of the peak and at the forest edge, throughout the moist forest, and the 

rocky summits. As a result of these visits the number of species identified from Mabu is 

currently 203 (see Annex 8). It is believed that there are more species to be recorded with an 

estimated total of around 250 species, bringing the total in line with lists from neighbouring 

mountains such as Mt Mulanje 70 km away. 

Of the 203 species recorded four are regarded as new species (Baliochila sp. nov., Cymothoe 

sp. nov., Epamera sp. nov. and Leptomyrina (Gonatomyrina) sp. nov.), three are new 

subspecies (Papilio pelodurus subsp. nov., Baliochila woodi subsp. nov. and Neocoenyra 

bioculata subsp. nov.) and in addition, according to Congdon, Collins & Bayliss (2010), 35 

are new records for Mozambique (two were previously found on Mt Namuli). 

Species of interest 

Cymothoe sp. nov. ‒ The new species of montane Cymothoe, a member of the Cymothoe 

aurivillius group, was first collected by J. Bayliss on Mt Namuli in November 2007, Mt Mabu 

in September 2008 and Mt Inago in 2009 (Bayliss 2008, Congdon & Bampton 2009, Congdon 

et al. 2010, Figure 28). Males were commonly found but females proved elusive. Eggs were 

noted on Rawsonia lucida (Flacourtiaceae), and larvae from all three mountains were 

successfully raised to the adult stage. Males from Mabu are paler yellow than those from Mts 

Inago and Namuli, and the yellow areas of the Namuli males are heavily dusted with black. 

Nevertheless it appears that they are all races of the same species. Montane Cymothoe 

characteristically reside in forest clearings; they are relatively sedentary and generally do not 

move between forest patches if separated by a significant distance. 

According to DNA analysis (Van Velsen & S. Collins, pers comm.), this species lies far apart 

from all other East African montane species, probably diverging in the Miocene some 3.8 

million years ago. There is also some variation between samples from each mountain 

population.


Epamera sp. nov. ‒ A new species of Epamera was collected on the summit of Mabu in 

October 2010 by J. Bayliss and M. Hassan, having been previously found on Mt Namuli by C. 

Congdon in 2009. Unfortunately all specimens caught were female and the male is yet to be 

discovered. Females were observed to lay on Actinanthella menyharthii (Loranthaceae). 

There is a group of related species in Iolaus (Epamera), which includes I. silanus, I. nolaensis 

and I. helenae. Other members of this group have been found on the Upper Sangha River, 

Congo Republic, on the Usambara, Nguru and Udzungwa mountains in Tanzania, and on the 

Mafinga Mountains in N Malawi and neighbouring Zambia, as well as in coastal Tanzania. 

The discovery of the new species extends the range of the group southwards. 

Baliochila sp. nov. ‒ Two specimens of a new species of Baliochila were caught on Mabu in 

early June 2009 by C. Congdon, S. Collins and M. Hassan (Figure 29). The upperside 

resembles that of B. woodi from Mt Mulanje, while the underside is closer to that of the 

Tanzanian species. It is unusual to find two species in the same group in the same forest, but 

the new species was found at lower elevations in riverine forest compared to the new 

subspecies of B. woodi. The habits of the two are also different ‒ B. woodi males perch 

around open spaces on ridges in the forest and are relatively easily netted, while those of the 

new species circle high under the canopy. It is probable that the B. woodi from Mabu 

represents a new subspecies. 

Leptomyrina sp. nov. ‒ Leptomyrina (Gonatomyrina) handmani was described from S 

Malawi and there are records from N and NW Zambia, but these may refer to another species. 

The new species of Leptomyrina from Mabu, which was first caught on the summit of Mt 

Mabu in October 2008 by J. Bayliss and which is also found on Mts Namuli and Inago, is 

darker and with more rounded wings; the hindwing lacks the tornal attenuation of L. 

handmani. 

Fig. 27. Papilio ophidicephalus (JB) 

Fig. 28. Cymothoe sp. nov. (JB) 

Fig. 29. Baliochila sp. nov. (JB)


7.7 Molluscs 

Leaf-litter molluscs were opportunistically collected from woodland and moist forest habitats 

by Julian Bayliss & Malaika Sacranie, with a total of 6 species being identified by A.C. van 

Bruggen (Netherlands). Not surprisingly, most species found on this rapid survey belonged to 

the genus Gonaxis ‒ the two commonest being G. vengoensis, known from central 

Mozambique, and G. elongatus with a range from central Mozambique and adjoining parts of 

Zimbabwe to S Malawi. The total number of terrestrial snail species expected would be 

around 40‒50 (van Bruggen, pers. comm. 2012). 

Table 10. List of molluscs collected on Mt Mabu, 2005‒2009. 

Family / species 

Pomatiasidae 

Tropidophora insularis (Pfeiffer, 1852) var. nyasana (Smith, 1899) 

Streptaxidae 

Gonaxis vengoensis Connolly, 1922 

Gonaxis elongatus (Fulton, 1899) 

Gonaxis sp. 

Subulinidae 

Pseudoglessula (Kempioconcha) liederi (Von Martens, 1895) 

Urocyclidae 

Carinazingis regalis Van Bruggen & De Winter, 1990 

Helicarionidae 

Sitala jenynsi (Pfeiffer, 1845) 

Species of interest 

Apart from a single shell from Lengwe National Park in Malawi, Pseudoglessula 

(Kempioconcha) liederi is otherwise known only from its type locality in coastal SE Tanzania 

(SW Lindi). The Mt Mabu record bridges the range gap. Carinazingis regalis is a relatively 

new genus and species first collected and described from Mt Mulanje. It is a very peculiar 

snail with a coronet between the tentacles on the head, first described from the Mt Mulanje 

complex at 1000‒2000 m. Congeneric specimens were reported from Ntchisi Mt, and shells 

probably belonging to this species are recorded from the Zomba and Nyika Plateaux in 

Malawi. Tropidophora insularis var. nyasana, with a large turbinate shell and spiral structure, 

was collected from woodland below 1000 m. Tropidophora species, widely distributed in 

southern and East Africa from the coast westward, although not really reaching Central 

Africa, are difficult to identify properly because of variation in the shell characters. They are 

tree dwellers usually found on tree stems in the forest. Sitala jenynsi is a widely distributed 

woodland species in coastal East Africa as far north as Kenya, reaching its southern limits in 

E Zimbabwe/C Mozambique. 

7.8 Freshwater Crustacea 

Six specimens of freshwater crabs from two of the rocky mountain streams near to the main 

forest campsite were opportunistically collected in October 2008. These were identified as 

Potamonautes choloensis (S. Daniels, pers. comm.), a species listed as Vulnerable on the 

2010 IUCN Red List. This is a range extension and the first record from Mozambique. 

As part of a broader study on Eastern and Southern African freshwater crabs, two new species 

were also described from Mt Namuli and Mt Mulanje (Daniels & Bayliss 2012), but not from 

Mt Mabu.


7.9 Biogeography 

Within most taxonomic groups that have been even partially surveyed, there is evidence of a 

significant biogeographical link with mountains to the north (in Tanzania) and west (Malawi), 

such as the Eastern Arc Mountains and Moreau’s Tanganyika‒Nyasa Montane Chain. 

However, there is more of an influence from the latter, coupled with significant elements from 

the mountains in northern Malawi and southern Tanzania. Biogegraphical aspects will be 

expanded upon in a forthcoming paper (Bayliss et al., in prep.). 

In addition, the number of endemic species from Mabu and surrounding mountains suggest a 

long period of isolation and ancient linkages with the north.


8. CONSERVATION 

The main conservation attributes of the Mt Mabu area are (a) the large extent and intact nature 

of the medium-altitude moist forest found here, and (b) the new species of vertebrates and 

invertebrates discovered. It is these attributes, along with the fact that it was previously almost 

unknown, that have given rise to much of the public interest in the area and make it unique in 

the Mozambique and regional context. And it is these attributes that should be the main 

objectives of conservation action. 

In this section we outline the main conservation issues and threats to Mabu's biodiversity and 

suggest some possible approaches and actions. Mabu's ornithological significance is also 

discussed by Dowsett-Lemaire (2010). 

8.1 Conservation Threats 

Although there are no significant immediate threats to the forest of Mt Mabu, a number of 

medium-term threats to its biodiversity and ecological integrity have been identified. These 

include, in order of assumed priority: (a) expansion of agriculture into lower parts of medium 

altitude forest if local populations or demand increases; (b) frequent fires are "hardening" the 

forest boundary and may, over a period of time, inhibit forest regeneration at the margins, 

resulting in a gradual diminution of the forest extent at the expense of moist woodland; (c) 

there is a slight danger of logging for timber, particularly in the surrounding moist woodland 

and in lower part of forest; and (d) the unsustainable level of bushmeat hunting in the forest 

by the local population. These are expanded upon below. 

a) Clearance for agriculture 

Inside the forest a few smaller patches cleared for subsistence agriculture were found on the 

eastern slopes, for example one near the forest fly camp (16 o 17'07"S, 36 o 23'13"S, 1325 m) 

where a small field of around 0.3 ha had been cleared perhaps 20‒25 years ago. This patch 

was now overgrown with secondary vegetation including Rubus pinnatus and many lianas. 

According to a local hunter, Ofelio Kavaliyawo, it was the locality where a few families took 

refuge during the civil war in the 1980s, living in lean-to shelters by a stream. Similar areas 

where there had been small fields in forest clearings were seen elsewhere. However, the 

ecological footprint appears rather small, in some respects similar to that of larger tree-fall 

sites. 

On the 1965/69 aerial photos there appears to have been some minor clearance at the lower 

forest margins on the southern and north-eastern slopes, suggesting previous clearance for 

small fields. But it could be that these areas were actually in woodland, not forest. Similar 

patches were not seen on recent satellite imagery, but with the resolution available it is not 

possible to confirm this. 

Also on the 1965 aerial photos two commercial tea estates be clearly be seen on the lower 

southern and south-eastern slopes, including the development of a network of tracks, and 

what looks like rural or agricultural development off part of the north-eastern slope. No 

evidence was seen that these tea plantations had been planted on what was forest, except 

perhaps across strips of what was riverine forest, although such evidence may not be apparent 

given that the plantations were put in perhaps 50 years ago. However, it is unlikely, due to 

altitude and limited moisture in the dry season, that there was much forest at these lower 

altitudes anyway, except strips of what here has been called lowland riparian forest, confined 

to bands along drainage lines with additional moisture.


Today what is seen on Mt Mabu is subsistence agriculture coming up to the forest margin, 

particularly on the south-eastern side where there appears to be more local population than 

elsewhere. Fire (see below) is hardening these boundaries. However, there was no evidence 

seen of encroachment of agricultural clearance into the forest itself ‒ clearance and cultivation 

comes up to forest boundary but appears to be in what was moist woodland or forest fringing 

vegetation. The woodland trees are felled and the dry, cleared vegetation is burnt in the dry 

season, the fire going uphill to the moist forest margin where it stops (Figure 12). Evidence 

was seen in the form of old field furrows that fallow areas were reused some years later, but it 

is not known what the fallow period is. 

However, with an increasing population and with the increasing demands of a cash economy, 

it is very possible that agricultural land will become scarce and land use pressures build up. 

This may well result in the forest margins being pushed back further and more forest being 

cleared. But on the mid-slopes of the mountain itself there were very few areas that would be 

level enough or sufficiently non-rugged to allow for small fields or settlements. Any 

agriculture or production would be very marginal. 

b) Fire 

Wildfires seemed to be very frequent during the 2008 dry season, with smoke being seen all 

around. At times, smoke and ash could be seen in the canopy right inside the forest. It was not 

the forest vegetation that was burning, and very little evidence was seen of such an 

occurrence, but the surrounding woodland and grassland (Figure 30). 

The fires are set to clear fallow agricultural fields or areas that have recently been cleared, and 

presumably fire has been used in such away for perhaps hundreds of years. What appears to 

be happening now, though, is that fire is increasingly frequent in any given locality, not just 

occasional and fierce. Wildfires come right up to the forest boundary, which they seem to 

"harden" in the sense that regeneration of forest species is greatly inhibited. However, there 

was no evidence seen to suggest that moist forest could or had extended much lower down the 

slopes from its present position, at least on the south-eastern side. The forest boundary is 

likely to be mostly determined by altitude/microclimate and dry season moisture availability, 

both in the soil and atmosphere in terms of low cloud, dew, etc. 

Fig. 30. Woodland clearance for agriculture and fire (JB).


c) Logging 

An important feature of the forests on Mt Mabu is that there was no evidence of logging or 

tree cutting seen. Although not 'pristine', the forest is remarkably intact and undisturbed. In 

many forests in Mozambique, for example those on Mt Namuli (Timberlake et al. 2009) or 

the coastal forests of Cabo Delgado Province (Timberlake et al. 2011), selective logging has 

had a significant impact on forest extent, structure and composition. This is not the case, as 

far as we could determine, on Mabu. 

There still remains a minor danger of commercial logging, but most trees found in the forest 

are not of desired species and in addition extraction would be difficult and expensive in such 

terrain. However, significant levels of illegal logging have been reported from parts of 

Zambézia Province closer to the coast (Mackenzie 2006, Mackenzie & Ribeiro 2009), so the 

organisation and logistics for extraction are available. 

Many trees of Pterocarpus angolensis were seen on a ridge felled and burnt in the process of 

clearing for fields, but interestingly there was no attempt to use the wood. 

d) Hunting 

According to a local hunter, mammal populations are now relatively low (see Section 7.2), 

although Blue Monkeys are still relatively common. It is only the largest mammal species that 

appear to have been extirpated. There are no or very few predators present. These losses are 

probably due to bushmeat hunting. Game birds such as Francolin and Guinea Fowl are still 

actively hunted, along with a number of small antelope. 

The forest is used by local hunters from the villages below, particularly the area around the 

summit. One trapped and dying klipspringer was seen here. The hunting method of choice 

appears to be gin-traps, a kind of spring trap (Figures 31, 33). Of the gin-traps seen, many 

were relatively small, designed for duikers or klipspringer, but could easily damage a human 

foot or leg. One very large and active trap was seen in 2006 (Figure 34) which appears 

designed to trap much larger animals. Also noted was a technique where a low drift fence of 

scrub is laid across a marked path, with a single 'gateway' through it. In the soil under this 

gateway a gin trap is set, attached to a strong nearby bush to stop any captured animal running 

away. Another technique designed, peresumably for elephant shrews and similar-sized 

rodents, is the construction of a low fence of small stakes. In a gap a small sprung snare is 

made using bark cord, which is triggered by the animal passing through (Figure 32). 

Fig. 31. Sun squirrel caught in gin trap, Mt Mabu (JB). 

Fig. 32. Small mammal trap and fence, 

Mt Mabu (JB).


Fig. 33 a,b. Small but powerful gin trap, 

Mabu summit (JT). 

Fig. 34. Large gin trap with Hassam Patel, 

Mabu forest (JB). 

Only a few villagers are hunters, a skilled ocupation that seems to run in families. Hunting 

trips can take place over a number of days in the forest while visiting set traps. Each hunter is 

aware of where the traps are, and who they belong to. 

The reduced number of desired animals, such as antelope, is acknowledged by the hunters. 

The present level of extraction is thought by us to be unsustainable, although could be 

maintained at a lower level. 

8.2 Protection and Conservation 

At present Mt Mabu is under no form of official protection, but as there are few immediate 

threats this is not a major problem. However, there are a number of medium-term threats and 

it will be necessary in due course to gazette the area as some sort of Reserve. Moves towards 

this are apparently being made at Provincial level (October 2012). 

It is not yet clear which would be the most appropriate form of protection under the new 

national Protected Area legislation. This legislation is now in the process of being finalised 

and appropriate structures developed for its implementation. A possible status for an area of 

the biodiversity significance and size of Mabu might be that of a Forest Reserve (under the 

National Directorate of Lands and Forests, NDLF, of the Ministry of Agriculture) or Reserva 

Especial (under the National Directorate for Conservation Areas, DNAC, of the Ministry of 

Tourism). An alternative would be some sort of Community Reserve that is 'owned' and 

managed by the local community. 

In 2011, the international conservation NGO, Fauna and Flora International (FFI), agreed to 

initiate and fund a community-based conservation project in the Mabu area. At the time of 

writing, this project is still being developed, although local NGO partners have been 

identified.


We do not wish to speculate here as to what are the most appropriate conservation actions. 

However, various suggestions for conservation action include: (a) limited ecotourism based 

and run by local communities, with trained local guides taking visitors through the forest to 

the peaks and able to explain something of the history and natural history of the area; (b) 

improvement and intensification of local agriculture to reduce the need to repeatedly clear 

new areas; (c) controlled, sustainable use of forest for some products (e.g. medicinal plants, 

some bush meat) is possible, but this would have to be limited in extent and carefully 

monitored to stop damage. A great attribute of the Mabu forests is their "pristine" nature ‒ not 

untouched or truly pristine, but fully intact and relatively undisturbed, self-regenerating 

without compromise. This needs to be retained. 

Fig. 35. Villagers at Mt Mabu forest camp (JB).


9. MAIN FINDINGS 

1. The moist forest present on Mt Mabu is around 7880 ha in extent, 5270 ha of which is 

thought to be at medium-altitude (1000‒1400 m). Forests at this altitude are increasingly 

scarce across the region as many similar areas have been cleared for agriculture. The 

Mabu forests are possibly the largest extent in southern Africa. In addition, the forests 

seen here have been little disturbed and are in very good condition. 

2. Although not pristine, the forest is intact and in very good ecological condition. No 

evidence of tree cutting was seen. 

3. Plant species composition is typical of medium altitude forest and inselbergs, although 

twelve new records for Mozambique and one (possibly two) new species were 

discovered. The species richness found (249 taxa) above 800 m altitude is normal for 

such areas. 

4. The finds of vertebrates were particularly exciting. Out of 15 reptile species recorded, 

three are new to science, including a forest viper and a chameleon, with a further three 

possibly new. A total of 126 bird species were recorded, including Southern Banded 

Snake Eagle, Green Barbet (belcheri race), Spotted Ground Thrush, Cholo Alethe, 

Swynnerton's Robin, Gunning's Akalat, Namuli Apalis and Dapple-throat. Several of 

these records represent significant extensions of known range. 

5. Large mammal populations are rather low, in part owing to bushmeat hunting. Although 

only a few species of small mammal were recorded, these include three new 

Mozambique records and one (possibly two) new species of bat. 

6. A particularly high number of butterflies (203) were recorded, including 7 species or 

subspecies new to science plus 35 other new Mozambique records. 

7. In conclusion, although the immediate threats to the forests on Mt Mabu are not as great 

as those on many other forests in Mozambique, with increasing land pressures they are 

likely to become much more significant. Two important threats are the high and 

unsustainable hunting for bushmeat, and the recent increase in fire incidence along the 

forest margins. Some sort of protected status will possibly be needed, along with a 

locally-based conservation project.


10. RECOMMENDATIONS 

Research 

1. Further documentation of the biodiversity of Mt Mabu is required, especially that of the 

medium-altitude moist forest areas. The number of new species and records of 

vertebrates and invertebrates is high, and many more are likely to be found. These appear 

to be primarily located in forests at medium altitude, a habitat that in many other 

localities across the region has been cleared. 

2. There should be a more detailed plant survey by habitat and altitude. Plant collecting to 

date has not been sufficiently comprehensive or systematic. Few new species of plant are 

likely to be found, but the number of new plant records is likely to increase once the 

northern and western sides of the mountain are explored. 

3. There is a need for better documentation of the extent of forest around the whole massif, 

and better typification of the range of forest habitats. In addition, investigation is required 

into what factors might be determining forest extent. 

4. An important next stage of investigation should be a comparative study of the various 

inselbergs across this part of northern Mozambique and southern Malawi. This would 

focus on determining similarities and differences between the various massifs in this 

region, and with those further afield in southern Tanzania and eastern Zimbabwe, 

especially the distribution ranges of endemics. It may be appropriate to define and 

describe an south-central African inselberg ecoregion. 

Management 

5. It is not yet clear which would be the most appropriate form of protection under the new 

national Protected Area legislation, possibly that of Forest Reserve (under the National 

Directorate of Lands and Forests, NDLF, of the Ministry of Agriculture) or Reserva 

Especial (under the National Directorate for Conservation Areas, DNAC, of the Ministry 

of Tourism). An alternative would be some sort of Community Reserve that is 'owned' 

and managed by the local community. Although not urgent, any longer-term future of the 

Mt Mabu forests will involve some sort of protected status. 

7. The development of low-impact tourism, based and run by local communities around the 

mountain, perhaps through a development NGO, is considered the most appropriate land 

use for the massif. Some regulation of external visitors may become necessary to ensure 

that benefit goes to these communities as well as to operators or other external partners, 

but without destroying the visitor's sense of adventure and excitement. 

8. A local guide programme should be developed, including training and licensing of a few 

local guides who would accompany each visitor group. These guides should be versed in 

local history and have a good knowledge of the forest, the plants and animals found 

there, and an ability to communicate this for ecotourism. 

9. Current levels of hunting and bush meat extraction need to be reduced so that antelope 

populations, and that of other medium-size animals, can increase. The present level is 

thought to be so high that populations of some species may become locally extinct.


10. There is a need for public education across the area aimed at reducing the incidence of 

uncontrolled wildfires. At present there is a lot of unnecessary burning in order to clear 

agricultural fields, with much collateral damage to forest and woodland biodiversity. 

11. Any ecotourism programme needs to have a base accessible by road, perhaps on the old 

Madal tea estate. This should feature in any local infrastructure planning. Appropriate 

"base" accommodation will also be required for ecotourism, the development of which 

offers an opportunity to local enterprise. 

12. The owners of the old tea estate on the south-eastern slopes, previously Grupo Madal, 

should be fully involved in the protection and development of Mabu forest, including 

ecotourism. There is some infrastructure there which could be rehabilitated, and there 

were also plans to re-open or replace the plantations. As well as the local population, the 

estate has long-established rights in the area.


11. ACKNOWLEDGEMENTS 

Firstly we wish to thank all members of the main expedition, and others who participated in 

smaller subsequent trips or contributed to the organisation and logistics. In particular, Carl 

Bruessow (MMCT), Paul Smith (MSB Kew) and Ofelio Kavaliyawo (local guide and hunter, 

Nangaze community). Plant identifications at the Herbarium, Kew were greatly assisted by 

David Goyder, Frances Crawford, Iain Darbyshire, Kaj Vollesen, Roger Polhill, Mike Lock, 

Tom Cope, Aaron Davis and Phil Cribb. 

Additional contributors to sections of this report were: 

Susana Baena (forest extent), GIS Unit, Royal Botanic Gardens, Kew, London. 

Ivan Bampton (butterflies), c/o African Butterfly Research Institute, Nairobi, Kenya. 

[deceased] 

Savel Daniels, (crabs), Dept. Botany & Zoology, University of Stellenbosch, South Africa. 

Ara Monadjem (bats), Dept. Biological Sciences, University of Swaziland, Swaziland. 

Jose Monteiro (forest extent), Estação Florestal de Mandonge, IIAM, Sussundenga, 

Mozambique. 

Stephen Mphamba (plants), Forest Research Institute of Malawi, Zomba, Malawi. 

Hassam Patel (plants), Mulanje Mountain Conservation Trust, Mulanje, Malawi. 

Claire Spottiswoode (ornithology), Dept. of Zoology, University of Cambridge, UK. 

Peter Taylor (small mammals), Durban Natural Science Museum, South Africa. 

Dolf Van Bruggen (molluscs), Naturalis Biodiversity Centre, Leiden, Netherlands.


12. BIBLIOGRAPHY & REFERENCES 

Ackery, P.R., Smith, C.R., Vane-Wright, R.I. (1995). Carcasson's African Butterflies: An 

annotated Catalogue of the Papilionoidea and Hesperioidea of the Afrotropical Region. 

British Museum (Natural History), London. 

Barbosa, L.A.G. (1952). Esboço da Vegetação da Zambézia. Separate from Documentário 

Moçambique No. 69. Centro de Investigação Científica Algodoeira, Lourenço Marques. 

Bayliss, J. (2008). Danger and discoveries in northern Mozambique. Lepsoc News Africa 

4(2008): 3–6. 

Bayliss, J. (2009). Trip Report, Mount Mabu Expedition, 10–30 October 2008. Internal 

report. MMCT, Mulanje. 

Bayliss, J., Monteiro, J., Fishpool, L., Congdon, C., Bampton, I., Bruessow, C., Matimele, H., 

Banze, A. & Timberlake, J. (2010). Biodiversity and conservation of Mount Inago, 

Mozambique. Report produced under the Darwin Initiative Award 15/036. Royal Botanic 

Gardens, Kew. 32 pp. Available at http://www.kew.org/ucm/groups/public/documents/ 

document/kppcont_046103.pdf 

Bayliss, J., Timberlake, J.R., Branch, W.R., Bruessow, C., Collins, S., Congdon, C., Michael 

Curran, de Sousa, C., Dowsett, R.J., Dowsett-Lemaire, F., Fishpool, L.D.C., Harris, T., 

Georgiardis, S., Kopp, K., Liggitt, B., Monadjem, A., Patel, H., Ribeiro, D., 

Spottiswoode, C., Taylor, P., Willcock, S. & Smith, P. (2012, in prep.). The discovery, 

biodiversity and conservation of Mabu forest – the largest mid-altitude rainforest in 

southern Africa. Oryx. 

Benson, C.W. (1950). A collection from Chiperoni Mountain, Portuguese East Africa. 

Bulletin British Ornithological Club 70: 51. 

BirdLife International (2008). Threatened Birds of the World 2008, CD-ROM. BirdLife, 

Cambridge. 

Branch, W.R. (2011). Treasure Mountain. Africa Geographic, March 2011: 30‒33. 

Branch, W.R. & Bayliss, J. (2009). A new species of Atheris (Serpentes: Viperidae) from 

northern Mozambique. Zootaxa 2113: 41–54. 

Branch, W.R. & Tolley, K.A. (2010). A new species of chameleon (Sauria: Chamaeleonidae: 

Nadzikambia) from Mount Mabu, central Mozambique. African Journal of Herpetology 

59(2): 157–172. 

Branch, W.R., Bayliss, J. & Tolley, K.A. (2012, in prep.). Pygmy chameleons of the 

Rhampholeon platyceps complex: The first records of Rhampholeon chapmanorum from 

Mozambique, the status of Brookesia platyceps carri Loveridge 1953, and the description 

of three new species from montane isolates in central Mozambique. 

Cabral, A. (2000). Borboletas de Moçambique. Natural History Museum(?), Maputo & 

Lisbon.


Congdon, T.C.E., Collins, S. & Bayliss, J. (2010). Butterflies of south east Africa's mountains 

(Mozambique and Malawi). Metamorphosis 21(2): 45‒107. 

Congdon, T.C.E. & Bampton, I. (2009). Musings from Mount Mabu. Metamorphosis 20(2): 

67–71. 

Curran, M. & Kopp, M. (2009). New records of bats (Order Chiroptera) from Mount Mabu, 

Mozambique. Unpublished report. University of Basel, Switzerland. 5 pp. 

Curran, M., Kopp, M. Bayliss, J. & Fahr, J. (2012, in prep.). Rhinolophus craveni nov. sp., a 

new species of horseshoe bat (Chiroptera: Rhinolophidae) from Mozambique. Acta 

Chiropterologica. 

D'Abrera, B.L. (1980). Butterflies of the Afrotropical Region. Lansdowne Editions, 

Melbourne. 

Daniels, S.R. & Bayliss, J. (2012). Neglected refugia of biodiversity: Mountainous regions in 

Mozambique and Malawi yield two novel freshwater crab species (Potamonautidae: 

Potamonautes). Zoological Journal of the Linnean Society 164: 498‒509. 

Dowsett, R.J. & Dowsett-Lemaire, F. (1984). Breeding and moult cycles of some montane 

forest birds in south-central Africa. Rev. Ecol. (Terre et Vie) 39: 89–111. 

Dowsett-Lemaire, F. (1989a). The flora and phytogeography of the evergereen forests of 

Malawi. I. Afromontane and mid-altitude forests. Bulletin du Jardin Botanique National 

de Belgique 59: 3–131. 

Dowsett-Lemaire, F. (1989b). Ecological and biogeographical aspects of forest bird 

communities in Malawi. Scopus 13: 1–80. 

Dowsett-Lemaire, F. (2008). Survey of birds on Namuli Mountain (Mozambique), November 

2007, with notes on vegetation and mammals. Dowsett-Lemaire Misc. Report 60. 26 pp. 

Available at http://ftp.rbgkew.org.uk/science/directory/projects/annex/namuli-birds- 

Dowsett.pdf 

Dowsett-Lemaire, F. (2010). Further ornithological exploration of Namuli and Mabu 

mountains (northern Mozambique), and the urgent need to conserve their forests. Bulletin 

of the African Bird Club 17(2): 159–177. 

Dowsett-Lemaire, F. & Dowsett, R.J. (2006). The Birds of Malawi: A handbook and atlas. 

Tauraco Press & Aves, Liège, Belgium. 

Dowsett-Lemaire, F. & Dowsett, R.J. (2009). The avifauna and forest vegetation of Mt. 

Mabu, northern Mozambique, with notes on mammals. Final report, October 2009. 

Unpublished report, Sumene, France. 20 pp. Available at http://www.kew.org/ 

ucm/groups/public/documents/document/kppcont_046100.pdf 

Fishpool, L.D.C. & Bayliss, J. (2010). Brief notes on the birds of Mount Inago, northern 

Mozambique. Bulletin of the African Bird Club 17(2): 198–201. 

Golding, J.S. [editor] (2002). Southern African Plant Red Data Lists. Soutern African 

Botanical Diversity Network Report no. 14. SABONET/SANBI, Pretoria.


Happold, D., Happold, M. & Hill, J. (1987). The bats of Malawi. Mammalia 51(3): 337‒414. 

Happold, M. & Happold, D.C.D. (1997). New records of bats (Chiroptera: Mammalia) from 

Malawi, east-central Africa, with an assessment of their status and conservation. Journal 

of Natural History 31: 805‒836. 

Harris, T., Darbyshire, I. & Polhill, R. (2011). New species and range extensions from Mt 

Namuli, Mt Mabu and Mt Chiperone in northern Mozambique. Kew Bulletin 66: 241– 

251. 

Hockey, P.A.R., Dean, W.R.J. & Ryan, P.G. [editors] (2005). Roberts ‒ Birds of Southern 

Africa (7th ed.). John Voelcker Bird Book Fund, Cape Town. 

Instituto Nacional de Geologia (1987). Carta Geológica, scale 1: 1 million. Instituto Nacional 

de Geologia, Maputo. 

Izidine, S. & Bandeira, S.O. (2002). Mozambique. In: Southern African Plant Red Data Lists 

(edited by J.S. Golding), pp. 43–60. SABONET, Pretoria. 

Johnson, W.P. (1884). Seven years' travels in the region east of Lake Nyasa. Proceedings of 

the Royal Geographical Society 6: 512‒533. 

Kalko, E. (1998). Organization and diversity of tropical bat communities through space and 

time. Zoology 101: 281‒297. 

Kassam, A.H., Van Velthuizen, H.T., Higgins, G.M., Christoforides, A., Voortman, R.L. & 

Spiers, B. (1981). Climatic data bank and length of growing period analysis. Field 

Document No. 33, Project MOZ/75/011, Assessment of Land Resources for Rainfed Crop 

Production in Mozambique. FAO, Rome. 

Kielland, J. (1990). Butterflies of Tanzania. Hill House, London. 

Last, J.T. (1887). On the Society's Expedition to the Namuli Hills, East Africa. Proceedings of 

the Royal Geographical Society 8: 467–490. 

Last, J.T. (1890). Mr J.T. Last's map of Eastern Africa, between the Rovuma and the 

Zambesi. Proceedings of the Royal Geographical Society 12: 223–224. 

Libert, M. (1999). Révision des genres Epitola Westwood, Hypophytala Clench et 

Stempfferia Jackson, et description de trois nouveaux genres (Lepidoptera Lycaenidae). 

A.B.R.I., Nairobi & Lambillonea, Tervuren, Belgium. 

Libert, M. (2004). Revision des Deudorix africains (Lepidoptera, Lycaenidae). A.B.R.I., 

Nairobi, Kenya & Lambillionea, Tervuren, Belgium. 

Mackenzie, C. (2006). Forest Governance in Zambézia, Mozambique: Chinese Takeaway! 

Final report for Fongza, Mozambique. 87 pp. Available at www.illegal-logging.info/ 

uploads/Mozambique_China.pdf. 

Mackenzie, C. & Ribeiro, D. (2009). Tristezas Tropicais: More sad stories from the forests of 

Zambézia. Justica Ambiental & ORAM, Maputo. 51 pp.


Maugham, R.C.F. (1910). Zambezia: A general description of the valley of the Zambezi River, 

from its delta to the River Aroangwa, with its hstory, agriculture, flora, fauna, and 

ethnography. John Murray, London. 

Ministério da Administração Estatal (2005). Perfil do Distrito de Lugela, Província da 

Zambézia. Serie Perfis Distritais. Consultancy report prepared by METIER Consultaria e 

Desenvolvimento, Maputo. 

Monadjem, A., Schoeman, M.C., Reside, A., Pio, D.V., Stoffberg, S., Bayliss, J., Cotterill, 

F.P.D., Curran, M., Kopp, M. & Taylor, P.J. (2010a). A recent inventory of the bats of 

Mozambique with documentation of seven new species for the country. Acta 

Chiropterologica 12: 371‒391. 

Monadjem, A., Taylor, P.J., Cotterill, F.P.D. & Schoeman, M.C. (2010b). Bats of Southern 

and Central Africa: A Biogeographic and Taxonomic Synthesis. Witwatersrand 

University Press, Johannesburg. 

Müller, T. (1999). The distribution, classification and conservation of rainforests in 

Zimbabwe. In: African Plants: Biodiversity, Taxonomy and Uses (edited by J. Timberlake 

& S. Kativu), pp. 221‒235. Royal Botanic Gardens, Kew. 

Müller, T. (2006). The distribution, classification and conservation of rainforests in Eastern 

Zimbabwe. Occasional Publications in Biodiversity No.19. Biodiversity Foundation for 

Africa/Zimbabwe Forestry Commission, Bulawayo. 

Müller, T., Mapaura, A., Wursten, B., Chapano, C., Ballings, P. & Wild, R. (2008). 

Vegetation survey of Mount Gorongosa. Occasional Publications in Biodiversity No.23. 

Biodiversity Foundation for Africa, Bulawayo. 

Musser, G.G. & Carleton, M.D. (2005). Superfamily Muroidea. In: Mammal Species of the 

World, 3rd edition (edited by D.E. Wilson & D.M. Reeder), pp. 894‒1531. John Hopkins 

University Press, Maryland, USA. 

Oatley, T.B. & Tinley, K.L. (1989). The forest avifauna of Gorongosa Mountain, 

Mozambique. Ostrich Suppl. 14: 57‒61. 

O'Neill, H.E. (1884). Journey from Mozambique to Lakes Shirwa and Amaramba. Part II, 

Exploration of the northern and north-eastern shores of lake Shirwa, and discovery of the 

lakes Amaramba and Chiuta, the true sources of the Lujenda River. Part III, Return 

journey from Lake Shirwa to the Mozambique coast at Angoche, November 1883 to 

January 1884. Proceedings of the Royal Geographical Society 6: 713–741. 

Parker, V. (2005). The Atlas of the Birds of Central Mozambique. Endangered Wildlife Trust 

& Avian Demography Unit, Johannesburg & Cape Town. 

Pedro, J.G. & Barbosa, L.A.G. (1955). A Vegetação. Esboço de Reconhecimento Ecólogica- 

Agricola de Moçambique. Centro de Investigação Científica Algodoeira, Lourenço 

Marques. 

Pringle, E.L.L., Henning, G.A. & Ball, J.B. (editors) (1994). Pennington's Butterflies of 

Southern Africa, second edition. Struik, Cape Town.


Rankin, D.J. (1885). Journey from Blantyre to Quillimane. Proceedings of the Royal 

Geographical Society 7: 655‒664. 

Reddy, S.J. (1984). General climate of Mozambique. Comunicação 19a, Série Terre e Agua. 

IIAM, Maputo. 

Ryan, P.G., Bento, C., Cohen, C., Graham, J., Parker, V. & Spottiswoode, C. (1999). The 

avifauna and conservation status of the Namuli Massif, northern Mozambique. Bird 

Conservation International 9: 315–331. 

Skinner, J.D. & Chimimba, C.T. (2005). The Mammals of the southern African Subregion. 

Cambridge University Press, Cape Town. 

Smithers, R.H.N. & Tello, J.L.P.L. (1976). Check List and Atlas of the Mammals of 

Moçambique. Museum Memoir No. 8. Trustees of the National Museums and 

Monuments of Rhodesia, Salisbury. 

Spottiswoode, C.N., Patel, I.H., Herrmann, E., Timberlake, J. & Bayliss, J. (2008). 

Threatened bird species on two little-known mountains (Chiperone and Mabu) in northern 

Mozambique. Ostrich 79: 1–7. 

Strugnell, A.M. (2006). A checklist of the spermatophytes of Mount Mulanje, Malawi. 

Scripta Botanica Belgica 34. National Botanic Garden, Meise, Belgium. 

Taylor, P.J., Stoffberg, S., Monadjem, A., Schoeman, M.C., Bayliss, J. & Cotterill, F.P.D. 

(2012). Four new bat species (Rhinolophus hildebrandtii complex) reflect Plio- 

Pleistiocene divergence of dwarfs and giants across an Afromontane archipelago. PLOS 

One 7(9): 1‒23 (e41744). 

Timberlake, J.R. & Shaw, P. [editors] (1994). Chirinda Forest: A Visitor's Guide. Forestry 

Commission, Harare, Zimbabwe. 

Timberlake, J.R. Golding, J.S. & Smith, P. (2006). A preliminary analysis of endemic and 

threatened plants of the Flora Zambesiaca area. In: Taxonomy and Ecology of African 

Plants and their conservation and sustainable use (edited by S.A. Ghazanfar & 

H.Beentje), pp. 749‒760. Royal Botanic Gardens. Kew. 

Timberlake, J.R., Bayliss, J., Alves, T., Baena, S., Francisco, J., Harris, T. & de Sousa, C. 

(2007). The biodiversity and conservation of Mount Chiperone, Mozambique. 

Unpublished report of Darwin Initiative project. Royal Botanic Gardens, Kew. 33 pp. 

Available at: http://www.kew.org/science-research-data/directory/projects/ 

DarwinMozambique.htm 

Timberlake, J.R., Dowsett-Lemaire, F., Bayliss, J., Alves, T., Baena, S., Bento, C., Cook, K., 

Francisco, J., Harris, T., Smith, P. & de Sousa, C. (2009). Mt Namuli, Mozambique: 

biodiversity and conservation. Report produced under the Darwin Initiative Award 

15/036. Royal Botanic Gardens, Kew. 115 pp. Available at http://www.kew.org/scienceresearch-data/directory/projects/DarwinMozambique.htm 

Timberlake, J.R., Goyder, D., Crawford, F., Burrows, J., Clarke, G.P., Luke, Q., Matimele, 

H., Müller, T., Pascal, O., de Sousa, C. & Alves, T. (2011). Coastal dry forests in northern 

Mozambique. Plant Ecology & Evolution 144: 126‒137.


Vincent, J. (1933a). The Namuli Mountains, Portuguese East Africa. The Geographical 

Journal 81: 314–327. 

Vincent, J. (1933b). [Four new species and eighteen new sub-species.... collected during the 

recent Portuguese East African Expedition]. Bulletin British Ornithological Club 53: 

129–149. 

Vincent, J. (1933–36). The birds of Northern Portuguese East Africa. Comprising a list of, 

and observations on, the collections made during the British Museum Expedition of 

1931–32. Ibis (13) 3: 611–652; 4: 126–160, 305–340, 495–527, 757–799; 5: 1–37, 355– 

397, 485–529, 707–762; 6: 48–125. 

Voortman, R. & Spiers, B. (1982). Land resources inventory, Mozambique. Map from Field 

Document No. 35, Project MOZ/75/011, Assessment of Land Resources for Rainfed Crop 

Production in Mozambique. INIA/FAO, Maputo. 

White, F. (1983). The Vegetation of Africa. Natural Resources Research No.20. UNESCO, 

Paris. 

White, F., Dowsett-Lemaire, F. & Chapman, J.D. (2001). Evergreen Forest Flora of Malawi. 

Royal Botanic Gardens Kew, London. 

Wild, H. & Barbosa, L.A.G. (1968). Vegetation map of the Flora Zambesiaca area. 

Supplement to Flora Zambesiaca. M.O. Collins, Harare. 

Williams, M.C. (2012). Afrotropical Butterflies & Skippers: A digital encyclopaedia, 11th 

edition. Available at http://www.atbutterflies.com 

Wilson, Smithett & Co. [compilers] (1962). Tea Estates in Africa. Wilson, Smithett & Co., 

London.

MT MABU, MOZAMBIQUE: - Royal Botanic Gardens, Kew

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?