Floristic richness in the Africa-Madagascar region: a brief history and prospective

Philippe MORAT Laboratoire de Phanérogamie , Muséum national d 'Histoire naturel le,

16 rue But ton, 75005 Paris, France. mora t@mnhn. f r

Porter P. LOWRY II Missour i Botanical Garden, P.O. Box 299, St. Louis, Missour i 63166-0299, U.S.A.

lowry@mobot .o rg Laboratoire de Phanérogamie , Muséum national d 'Histoire naturel le,

16 rue Buffon, 75005 Paris, France. lowry-p@mnhn. f r

K E Y W O R D S Africa,

M a d a g a s c a r , floristic r ichness,

floras.

A B S T R A C T The floras of southern Africa (22,000 species) and Mediterranean Africa (6,000 species) are comparatively well known and have been the subject of many recent treatments; however, numerous gaps still exist in our understan­ding of the flora of tropical Africa (25,000 species). Certain regions are not covered by any of the current flora or inventory projects, whereas others are inaccessible for political reasons. Data on the floristic richness of the coun­tries and regions in the Africa-Madagascar area are updated. Using the example of Madagascar, the density index is re-evaluated, showing that the values pteviously presented are underestimates. Most of the current floras and other large projects are progressing very slowly because of political insta­bility in the areas concerned, lack of sufficient funding, and sepecially a shor­tage of qualified taxonomists. Several checklists are being prepared using modern computer technology, which now makes it possible to envision both a global checklist fot the entire Africa-Madagascar area, and a General Flora of Tropical Africa similar to the Flora Neotropica, as suggested by LfiONARD in 1975.

sér. 3 • 1997 • 19 (1 ) 101

M o r a t Ph. & Lowry II P .P .

M O T S C L É S Afr ique ,

M a d a g a s c a r , r ichesse f lor is t ique,

flores.

R É S U M É Si la flore de l'Afrique australe (22 000 espèces) et celle de l'Afrique méditet-

ranéenne (6 000 espèces) sont relativement bien connues et sont l'objet de

nombreux ttaitements récents, la flore tropicale africaine (25 000 espèces)

présente encore de nombreuses lacunes dans sa compréhension. Certaines

régions ne sont couvertes par aucun projet de Flore ni même d'inventaire,

tandis que d'autres sont rendues inaccessibles par des troubles d'ordre poli­

tique. Des données entièrement actualisées concernant la richesse floristique

dans différents états ou régions du domaine africano-malgache sont ici four­

nies. De même, à partir de l'exemple de Madagascar une ré-évaluation du

"density index" est entreprise montrant la sous-estimation admise jusqu'à

présent pour les pays du domaine africano-malgache. La plupart des Flores et

des grands projets actuels ne progressent que ttès lentement, par manque de

stabilité politique des états concernés, des crédits et surtout de taxonomistes

qualifiés. Grâce aux possibilités offertes par l'outil informatique, de nom­

breuses cheldists sont en cours d'élaboration qui permettent dès maintenant

d'envisager d'une part une checklist globale pour l'ensemble du domaine étu­

dié et d'autre part la réalisation d'une Flore Générale d'Afrique Tropicale, à

l'image de la Flora Neotropica, comme l'avait déjà proposé LÉONARD en

1975.

I N T R O D U C T I O N

T h e Africa-Madagascar phytogeographic region

covers approximately 3 0 . 9 mil l ion k m 2 , inclu­

ding numerous islands in the surrounding waters

(e.g., Mascarenes, Seychelles, Comoros , Socotra,

Bioko, C a p e Verde, etc.). According to the most

recent estimates, the region includes a vascular

plant flota o f about 5 5 , 0 0 0 species, with nearly

4 5 , 0 0 0 occurring on the African continent, an

estimated 2 1 - 2 5 , 0 0 0 o f which are found in the

tropical areas south o f the Sahara and north o f

N a m i b i a , Bo t swana , and S o u t h Africa (World

Conservation Moni tor ing Centre 1 9 9 2 ; B E E N T J E

et al. 1 9 9 4 ; L E B R U N & S T O R K 1 9 9 5 ) and about

2 2 , 0 0 0 in S o u t h e r n Afr ica ( C O W L I N G et a l .

1 9 8 9 ; S M I T H et al. 1 9 9 5 ) , with an additional 1 0 -

1 2 , 0 0 0 s p e c i e s o c c u r r i n g in M a d a g a s c a r

( H U M B E R T 1 9 5 9 ; P H I L L I P S O N 1 9 9 4 ; S C H A T Z et

al. 1 9 9 6 ) . While the flora o f this region is not

nearly as rich as those o f tropical America (with

ca. 9 0 , 0 0 0 species; H E N D E R S O N et al. 1 9 9 1 ) and

Asia (about 4 5 , 0 0 0 species), it is nevertheless far

from being well known.

T h e evolution and diversification o f the flora o f

the Africa-Madagascar region are largely due to

the age o f the cont inent and adjacent is lands,

and to their geological and climatological histo­

ries, which have together resulted in the very

b r o a d a r ray o f e d a p h i c c o n d i t i o n s o b s e r v e d

today. A tremendous range o f habitats are found

in the region, which are reflected in an impres­

sive variety o f vegetation types, from dense, per-

h u m i d f o r e s t s to d r i e r d e c i d u o u s f o r e s t s ,

shrublands, woodland and grassland, and some

of the world's harshest deserts, each o f which also

has one or more corresponding degraded forms

such as secondary grass lands ("savannas") and

secondary bushlands that are the result o f human

action.

T h e purpose o f this paper is to evaluate pro­

gress in the development o f our understanding o f

the botanical diversity in the Africa-Madagascar

region. T h e information presented here is inten­

ded not only to summarize what we know today

and where the gaps occur in our knowledge, but

also to st imulate further studies o f the flora o f

this important and intriguing region.

102 ADANSONIA, sér. 3 • 1997 • 19 (1 )

Flor is t ic r ichness in Af r i c a -Madagasca r

D I S T R I B U T I O N O F F L O R I S T I C R I C H N E S S

T h e floristic richness in the Africa-Madagascar region is very irregularly distributed. Early stu­dies, which were global in scale, pointed out the existence o f a center o f exceptional diversity in the C a p e Province, which was quickly recognized as a floristic k ingdom ( E N G L E R & D l E L S 1 9 3 6 ) ,

an interpretation that has consistently been fol­lowed by modern authors (e.g., T U R R I L L 1 9 5 9 ; G O O D 1 9 7 4 ; T A K H T A J A N 1 9 8 6 ) . Similarly, the

Malagasy floristic region has long been recogni­zed as hav ing remarkab ly high levels o f bo th e n d e m i s m a n d spec i e s d ive r s i ty ( E N G L E R & D I E L S 1 9 3 6 ; H U M B E R T 1 9 5 9 ; K O E C H L I N et al.

1 9 7 4 ; T A K H T A J A N 1 9 8 6 ; P H I L L I P S O N 1 9 9 4 ) .

Many estimates o f the number o f species occur-

Fig 1.—Main phytochoria of Africa and Madagascar: I, Guineo-Congolian regional centre of endemism. II, Zambezian regional centre of endemism. Ill, Sudanian regional centre of endemism. IV, Somalia-Masai regional centre of endemism. V, Cape regional centre of endemism. VI, Karoo-Namib regional centre of endemism. VII, Mediterranean regional centre of endemism. VIII, Afromontane archi­pelago-like regional centre of endemism, including IX, Afroalpine archipelago-like region of extreme floristic impoverishment (not shown separately). X, Guinea-Congolia/Zambezia regional transition zone. XI, Guinea-Congolia/Sudania regional transition zone. XII, Lake Victoria regional mosaic. XIII, Zanzibar-lnhambane regional mosaic. XIV, Kalahari-Highveld regional transition zone. XV, Tonga land-Pondo land regional mosaic . XV I , Sahel regional t ransi t ion zone. XV I I , Sahara regional t ransi t ion zone . XVI I I , Mediterranean/Sahara regional transition zone. XIX, East Malagasy regional centre of endemism. XX, West Malagasy regional centre of endemism.—After WHITE 1993.

ADANSONIA, s e r . 3 • 1 9 9 7 • 1 9 ( 1 ) 103

M o r a t Ph . & L o w r y II P.P.

ring in the Africa-Madagascar region have been made over the years (see below), each o f which was the result o f an updated compilation o f data for various areas. These estimations have evolved constantly as readjustments were made following the completion o f floras, revisions o f individual famil ies , m o n o g r a p h s , checkl is t s , m a p s , field work, etc.

T h e floristic richness o f tropical Africa, inclu­d ing the area sourh o f the Saha ra to A n g o l a , Zambia , Z imbabwe, and Mozambique , was for m a n y years substant ia l ly overes t imated at 3 0 -3 5 , 0 0 0 species (e.g., B-RENAN 1 9 7 8 ) . Recent stu­d ie s by L E B R U N & S T O R K ( 1 9 9 5 ) i n d i c a t e ,

however, that a more realistic figure wou ld be a b o u t 2 5 , 0 0 0 spec ies , s l ight ly m o r e than the f igure o f 2 2 , 0 0 0 c u r r e n t l y r e p o r t e d for the Southern African region ( N a m i b i a , Botswana , S o u t h A f r i c a , L e s o t h o , a n d S w a z i l a n d ) ( C O W L I N G et al. 1 9 8 9 ; S M I T H et al. 1 9 9 5 ) , while

about 1 0 , 0 0 0 species are throught to occur in N o r t h Afr ica ( Q U E Z E L 1 9 8 5 ) . E s t i m a t e s for Madagascar range from 1 0 , 0 0 0 (HUMBERT 1 9 5 9 ; PHILLIPSON 1 9 9 4 ) to a p p r o x i m a t e l y 1 2 , 0 0 0 (GuiLLAUMET & KOECHLIN 1 9 7 1 ; S C H A T Z et al. 1 9 9 6 ) , the latter o f which is certainly more real­istic.

O n e o f the most widely cited publications that assesses the distributions o f flotistic richness in Africa is the A E T F A T / U N E S C O / U N S O vegeta­tion map ( W H I T E 1 9 8 3 ) . W H I T E recognized 2 0

phytochor ia , inc lud ing 1 0 regional centers o f endemism separated from each other by one o f s even r e g i o n a l t r a n s i t i o n z o n e s ; 2 r e g i o n a l mosaics and an afroalpine archipelago-like region o f extreme floristic impover i shment were also recognized. A center o f endemism was defined as having more than 5 0 % of its flowering plant spe­cies endemic to the region, and a total o f 1 , 0 0 0 endemic or near-endemic species ( W H I T E 1 9 8 3 , 1 9 9 3 ) . These regional centers o f endemism cor­respond largely with the floristic areas recognized by E N G L E R & D I E L S ( 1 9 3 6 ) and G O O D ( 1 9 7 4 ) ,

a n d s u b s e q u e n t l y s i m p l i f i e d by T A K H T A J A N ( 1 9 8 6 ) . W H I T E ( 1 9 8 3 ) provided an assessment o f

the total number o f species and the percentage o f e n d e m i s m in e a c h p h y t o c h o r i o n ( F i g . 1 ) , although in at least some cases these were proba­bly underestimates.

DISTRIBUTION BY COUNTRY

In otder to evaluate our current understanding of African floristics, an assessment can be made of the total number o f species thought to be pre­sent in each country, or each group of countries be ing treated together by o n g o i n g or recently completed flora projects. Mos t o f these projects (which include the publication o f comprehensive floras, bu t in s o m e cases also inventories and checklists) strictly follow political borders, which in nearly all cases are art if icial , at least wi th regard to the d i s t t i b u t i o n o f p lan t diversi ty. S o m e projects are strictly national in scope (e.g., F lore d u G a b o n , Flore d u C a m e r o u n ) , while others include several countries (such as Flore de l'Afrique Centrale, Flora o f Tropical East Africa, and the Numerat ion des plantes d'Afrique tropi-cale). There are only a few examples o f projects that cut across political boundaries, such as Flora Zambesiaca, which includes the Caprivi Strip of Namib ia , or the Flora o f West Tropical Africa, whose coverage has a northern limit that follows the sixteenth parallel, corresponding approxima­tely wi th the s o u t h e r n l imi t o f the F lo re de lAfr ique du N o r d and the Med-Checkl i s t pro­ject.

T h e current estimate o f the number o f species that occur in each count ry within the Africa-Madagascar region is shown in Fig. 2 . T h e data are based on L E B R U N (1976 , as revised in 1991) , the W o r l d C o n s e r v a t i o n M o n i t o r i n g C e n t r e ( 1 9 9 2 ) , a n d B E E N T J E et al. ( 1 9 9 4 ) , a n d have been further revised and updated where possible using a combinat ion o f more recently available informat ion and several older sources that are considered to provide the most accurate figures ava i l ab le t o d a y for pa r t i cu l a r areas ( G U I N E A 1 9 4 6 ; K E M P 1 9 8 3 ; H I E P K O & S C H O L Z 1 9 8 9 ;

A K P A G A N A & B O U C H E T 1994; A K P A G A N A et al.

1 9 9 4 ; F R I E D M A N N 1 9 9 4 ; L E O N A R D 1 9 9 4 ;

B O U L O S 1995 ; L E B R U N 1997a, 1997b) . T h e cur­rent data are not totally comparab le from one country or region to another, primarily because the documents on which they are based do not always indica te whether for e x a m p l e they are dealing with all vascular plants or angiosperms only, or whether they cover only the indigenous flota or all o f the species present in a given area,

104 ADANSONIA, ser. 3 • 1997 • 19 (1 )

Flor is t ic r ichness in Af r i ca -Madagasca r

Fig 2.—Distribution of floristic richness by country or ongoing project.

including those that are naturalized and/or culti­vated. Fur thermore, a l though the figures p u b ­lished for some countries are presented as being accurate and exhaustive down to the last species (such as values o f 2 , 4 2 3 species for T o g o , or

1 , 2 0 3 for Burkina Faso), in fact they only reflect the information available to the authors at the t ime their works were completed (e.g., G U I N E A 1 9 4 6 ; L E B R U N et al. 1 9 9 1 ) .

In reality, all o f the numbers presented in Fig. 2

ADANSONIA, s é r . 3 • 1 9 9 7 • 1 9 ( 1 ) 105

M o r a t Ph. & L o w r y II P.P.

must be regarded as approximations, which will inevitably change as additional botanical explora­tion, analysis o f specimens, and taxonomic revi­sion proceed. T h u s , while the level o f accuracy ind ica ted at the t ime o f pub l i ca t i on o f these works can reasonably be accepted, provided that the constraints and limitations indicated by the authors are kept in mind (such as poorly explo­red or unknown areas, and the lack o f sufficient herbarium collections), even the mos t carefully compi l ed figures nevertheless quickly b e c o m e o u t d a t e d as new col lec t ions and in fo rma t ion become available, and as taxonomic concepts are revised . For e x a m p l e , the n u m b e r o f spec ies recorded from Togo increased in just a decade from 2 ,195 reported by B R U N E L et al. (1984) to 2 ,423 cited by A K P A G A N A & G U E L L Y (1994; after corrections are made to the errors in the figures they provide).

For areas in which flora projects are still in pro­gress, the data presented in Fig. 2 ate estimates derived by extrapolation from the work already completed, taking into consideration the projec­ted richness o f the particular groups that remain to be treated. In most cases, these figures were either provided by, or at least verified by, the per­s o n ^ ) responsible for coordinating the respective

projects. In a few cases, we have had to estimate the numbers o f species ourselves using the avail­able base line information, much o f which is very difficult to verify.

It is wor th no t ing that the " D a h o m e y G a p " referred to by F R O D I N (1984) has now been clo­sed, or at least largely so, by the completion o f the "Flore analytique du T o g o " ( B R U N E L et al. 1984) , as well as a careful inventory o f the plants occurring in Burkina Faso ( L E B R U N et al. 1991) and a somewhat cursory listing o f those in Benin ( S O U Z A 1989) . O n the other hand, Liberia and Guinea remain very poorly known.

C U R R E N T S T A T E A N D P R O G R E S S O F F L O R A S

Thanks in large part to the Association for the Taxonomic Study of the Flora o f Tropical Africa (known by its French acronym, A E T F A T ) , pro­gress made by the various flora projects in sub-S a h a r a n Afr ica has been regu la r ly e v a l u a t e d ( B A M P S 1 9 9 4 ) . S ince the m o s t recent u p d a t e (i.e., between June 1994 and December 1996) , new treatments (or partial treatments) covering 65 families have been publ ished entirely or in

TABLE 1.—Progress in ongoing African flora projects between June 1994 and December 1996.

Carrier. Gabon C. Afr. Ethiopia Somalia FTEA Angola Zamb. S.Afr. Madag. Masc. TOTAL

Number of families published between June 1994 and December 1996 (according to BAMPS

1994 and updated) (pers. comm.)

0 1 2 1 1 32 12 2 0 0 13 65

Date of publication of first fascicule or volume

1963 1960 1948 1991 1993 1952 1937 1960 1963 1936 1976

Percent published or accessible

25 33 67 13 46 72 55 45 13 60 65

Camer. = Flore du Cameroun Gabon = Flore du Gabon C. Afr. = Flore de TAfrique Centrale Ethiopia = Flora of Ethiopia Somalia = Flora of Somalia FTEA = Flora of Tropical East Africa

Angola = Conspectus Florae Angolensis Zamb. = Flora Zambesiaca S. Afr. = Flora of Southern Africa Madag. = Flore de Madagascar et des Comores Masc. = Flore des Mascareignes

106 ADANSONIA, sé r . 3 • 1 9 9 7 • 1 9 ( 1 )

Flor is t ic r ichness in Af r i c a -Madagasca r

part in the ongoing floras, as indicated in Table 1. D a t a from regional or na t iona l floras (for example those for T o g o and M o z a m b i q u e ) , as well as from more restricted or incomplete works (e.g. , floras l imi ted to specific g roups such as trees, ferns, grasses, etc.) or incomplete projects ( i n c l u d i n g t h o s e for G h a n a , N i g e r i a , I vo ry Coast , and Rwanda) , have only been taken into considerat ion in a few cases, primarily because these treatments cover areas that are included in other, more extensive projects such as the Flora o f West Tropical Africa, Flora Zambesiaca, Flore d'Afrique Centrale, etc.

Careful examina t ion o f the list o f 6 5 family treatments presented in Table 1 shows that they are very unequal in size, largely reflecting the overall richness o f each group within the Africa-M a d a g a s c a r region. S o m e addi t ional variat ion may also be due to differences in species concepts o r t a x o n o m i c s y s t e m ( e . g . , r e c o g n i t i o n o f Trist ichaceae as distinct from Podos temaceae) . Nevertheless , as indicated by B A M P S , T H U L I N ,

P O P E and W H I T E H O U S E (pers. c o m m . ) , several very large and important groups have been re­v i s e d s ince B A M P S ' m o s t recent c o m p i l a t i o n (1994 ) , including:

— P o a c e a e ( F l o r a o f E t h i o p i a , F l o r a o f Somal ia) ,

— Euphorbiaceae, in part (Flora Zambesiaca , Flore de l'Afrique Centrale),

— Orch idaceae , all (Flora o f Somal i a ) or in part (Flora Zambesiaca) .

It is perhaps worth not ing that two thirds o f the family treatments published during the per­iod f rom 1 9 9 4 to the end o f 1 9 9 6 (Table 1) appeared in the Flore des Mascareignes and Flora of Somalia , which both cover a small area with a limited number o f species, the latter o f which is a rapidly expanding project.

Regarding the current state o f progress o f the various sub-Saharan flora projects and floristic compi la t ions , three have now been completed: Flora o f West Tropical Africa in 1972 ; Flora o f Sudan in 1956; and the Prodromus einer Flora von Siidwestafrika in 1 9 7 2 . Th ree other large endeavor s are n o w a p p r o x i m a t e l y two th i rds comple te : Flora Zambes iaca , Flora o f Tropical East Africa, and Flore de l'Afrique Centrale. T h e other o n g o i n g sub -Saha ran projects , cover ing

G a b o n , C a m e r o o n , M a d a g a s c a r a n d t h e C o m o r o s , Soma l i a , E th iop ia , the Masca renes , etc., are at various stages o f completion, and are in m a n y cases progress ing very slowly, usually d u e to a c o m b i n a t i o n o f i n su f f i c i en t funds and/or a lack o f personnel.

In Southern Africa, the Flora Capensis is now outdated, and the Flora o f Southern Africa, cov­e r i n g t h e a r e a s o u t h o f A n g o l a , Z a m b i a , Z imbabwe and Mozambique , is only about 1 3 % comple te (P. G O L D B L A T T , pers. c o m m . ) and is currently inactive, a l though it remains a long-term priority project ( S M I T H et al. 1995) . Th i s area is considered one o f the best known floristi-cally within the Africa-Madagascar region, large­ly because numerous modern partial floras and monographs have been published covering all or parts o f Southern African. Also, a comprehensive d a t a b a s e , P R E C I S , has b e e n c o m p i l e d tha t inc ludes spec imen a n d taxon in fo rma t ion on essentially all o f the estimated 2 2 , 0 0 0 species and 2 , 5 0 0 infraspecific taxa occur r ing in the area ( S M I T H et al. 1995) .

In Nor th Africa, a flora is available for Egypt ( T A C K H O L M 1 9 6 9 , 1 9 7 4 ) , upda ted by a more recent checklist ( B O U L O S 1995) , while the rest o f the region, including the countries from Libya to the northern part o f Mauri tania , is covered by M A I R E ' s F lo re de l 'Af r ique d u N o r d ( 1 9 5 2 -1 9 8 7 ) , w h i c h has b e e n u p d a t e d by Q j J E Z E L ( 1 9 8 5 ) , and is now 6 5 % comple te . T h e M e d -Checklist O P T I M A project, which includes the M e d i t e r r a n e a n pa r t o f Af r i ca , has c o m p i l e d in format ion on a b o u t 6 , 0 0 0 o f the es t imated total o f 2 0 , 0 0 0 species that occur in the area, but is now progressing very slowly for the reasons mentioned above regarding other projects.

P R O B L E M S A N D I N F O R M A T I O N G A P S

In addi t ion to the floristic works men t ioned above, numerous monographs, distribution maps a n d i n d e x e s a re a l s o a v a i l a b l e , i n c l u d i n g Dis t r ibu t iones p l an t a rum afr icanarum (Anon. 1969-) and the Index de repartition des plantes vasculaires d 'Afrique ( L E B R U N & S T O R K 1 9 7 7 ,

1 9 8 1 ; S T O R K & L E B R U N 1988) , along with many vegetation maps at various scales, all o f which are

ADANSONIA, ser. 3 • 1997 • 19 (1 ) 107

M o r a r Ph . & L o w r y II P.P.

Flor is t ic r ichness in Af r i c a -Madagasca r

too numerous to cite here. A number o f gaps never theless r ema in , p r imar i ly b e c a u s e s o m e parts o f the Africa-Madagascar region fall outside the areas treated by the ongoing or recently com­pleted flora projects. T h e most striking examples are C o n g o , the Centra l African Republ ic , and Chad , which are only covered by floristic listings whose accuracy and completeness are uncertain. Addi t ional gaps also exist because the available reference collections, upon which even the most advanced floristic works must be based, are far from adequate.

As o f 1 9 7 2 , for example, an area in C h a d com­prising about ten square degrees (that is, a hun­dred thousand k m 2 ) , and situated in a zone that is considered to be potentially rich in plant spe­c i e s , h a d p r a c t i c a l l y n e v e r b e e n c o l l e c t e d ( L E B R U N et al. 1 9 7 2 ) , and no addi t ional field work has been reported since then. Furthermore, many areas, or even entire countries, must still be regarded as " insuff ic ient ly known" , based on L E O N A R D ' S ( 1 9 6 8 ) s t anda rd o f three t imes as many specimens as species, which is very modest indeed.

Figure 3 shows the level o f floristic exploration in Africa as o f 1 9 6 5 (as presented by LfiONARD 1 9 6 8 ) , and Fig. 4 presents the same information 1 4 years later ( H E P P E R 1 9 7 9 ) . Both maps recog­nized areas that are: 1 ) poorly known, in which the number o f specimens collected is less than the number o f species thought to be present; 2 ) moderately well known, with the number o f spe­cimens ca. 1-3 times the number o f species; and 3 ) well known, where the avai lable spec imen base is greater than three t imes the es t imated number o f species. T h e botanical specimen base clearly increased between 1 9 6 5 and 1 9 7 9 for cer­tain regions o f Angola , East Africa and S u d a n ( w h i c h w e n t f r o m b e i n g " m o d e r a t e l y we l l known" to "well known"), and for a fairly large part o f Southern Africa and the western part o f the continent (from "poorly known" to "modera­te ly we l l k n o w n " ) . H o w e v e r , s eve ra l o t h e r regions, such as Ethiopia, Somal ia , Mauri tania, and Mali , appear to have regressed as a result o f an initial overestimate o f the number o f speci­mens available. Madagascar was only superficial­ly t r e a t ed by LEONARD ( 1 9 6 8 ) a n d H E P P E R ( 1 9 7 9 ) , and while botanical exploration there is

still far from uniform, substantial progress was m a d e between the t ime o f L E O N A R D ' S analysis and that o f H E P P E R .

These figures highlight the subjective nature o f this k ind o f evaluat ion, and indicate jus t how slow progress was dur ing the 1 4 years through 1 9 7 9 . Clearly a major effort has been made in many parts o f the Africa-Madagascar region since then, but even more will be required in the fu­ture, especially consider ing that in many areas r e g a r d e d as p o o r l y or o n l y m o d e r a t e l y wel l known botanically, the native vegetation is cur­rently undergoing rapid and accelerating levels o f d e g r a d a t i o n . H E P P E R ( in C A M P B E L L &

H A M M O N D 1 9 8 9 ) indicated that an es t imated 2 6 0 , 0 0 0 ha o f forest were being lost in Nigeria each year, and that 4 0 0 , 0 0 0 ha were destroyed annually in Ivory Coast . For Madagascar, G R E E N & S U S S M A N ( 1 9 9 0 ) estimate that 1 1 1 , 0 0 0 ha o f

rain forest are cut each year in the eastern part o f the country alone, as much as or more than in Amazonia.

P R A N C E ( 1 9 7 7 ) , V I N K ( 1 9 8 1 ) , and others have

proposed an alternative me thod for evaluating our level o f k n o w l e d g e o f f loras , the dens i ty index. According to these authors, a given area mus t have more than 1 0 0 collections per 1 0 0 k m 2 (or one per k m 2 ) to be regarded as suffi­ciently known. Using this method, C A M P B E L L & H A M M O N D ( 1 9 8 9 ) determined that no country in Africa reached this level as o f 1 9 8 1 (Table 2 ) , and that based on estimates o f the rate o f collec­ting at that t ime, they indicated that it would take several decades , centuries, or even longer, depending on the case (Table 3 ) , before the col­lection density would meet the criteria for being rega rded as suff ic ient ly k n o w n (F ig . 4 ) . T h e figures presented by C A M P B E L L & H A M M O N D ( 1 9 8 9 ) are, however, clearly underestimates for at least certain count r ies , a n d perhaps for all o f them. For example, the density index o f 9 assi­gned to Madagascar (Table 2 ) for both 1 9 6 4 and 1 9 8 1 , corresponding to just 5 3 , 1 0 0 numbers in all for the country, obviously does not take into account the large number o f collections (more than 1 0 , 0 0 0 ) made in the 1 7 intervening years. Moreover, based on a careful estimate made by the first author in 1 9 7 6 , nearly 2 0 0 , 0 0 0 numbers from Madagascar, representing a density index o f

ADANSONIA, s e r . 3 • 1 9 9 7 • 1 9 ( 1 ) 1 0 9

M o r a t Ph . & L o w r y II P.P.

TABLE 2.—Number of herbarium specimens per tropical country/region, 1 9 5 2 - 1 9 8 1 (from CAMPBELL & HAMMOND 1 9 8 9 ) .

Collection density index

Country 1952 1954 1956 1959 1964 1974 1981

Sierra Leone 0 0 0 0 0 4 1 5 0

Liberia 0 0 0 0 3 6 6

Ivory Coast 0 1 1 1 1 7 7

Ghana 0 1 0 1 0 2 1 2 1 4 2 4 8

Benin & Togo 0 0 0 0 0 0 5

Nigeria 2 3 3 3 4 6 1 4

Cameroon 0 0 1 1 1 1 1 0

Gabon 0 0 0 0 0 0 1

Congo Republic 0 1 1 1 3 9 9

Zaire 0 0 4 4 4 5 5

Central African Republic 0 1 1 1 1 1 1

Angola 0 0 1 1 1 4 5

Uganda 3 1 5 1 5 1 5 1 7 3 0 7

Kenya 1 4 1 7 1 7 1 7 2 8 5 5 6 4

Tanzania 0 0 0 1 1 3 5

Madagascar 2 2 2 2 9 9 9

Collection density index = number of col lect lons/100 k m 2 within the borders of the tropical-forested country.

TABLE 3.—Rates of botanical inventory in tropical countries/regions (from CAMPBELL & HAMMOND 1 9 8 9 ) .

No. years required to reach level of

Collections/ 100 specimens/ No. collections/ 100 km2/year 100 km 2 within

Country/Region 100 km 2 in 1981 1974-1981 country/region

Sierra Leone 5 0 1 . 1 8 4 3

Liberia 6 0 . 0 0 OC

Ivory Coast 7 0 . 0 0 CO

Ghana 4 8 0 . 9 0 5 8

Benin & Togo 5 0 . 7 5 1 2 6

Nigeria 1 4 0 . 4 9 1 7 5

Cameroon 1 0 1 . 3 8 6 5

Gabon 1 0 . 1 6 6 1 8

Congo Republic 9 0 . 0 0 OO

Zaire 5 0 . 0 0 oo

Central African Republic 1 0 . 0 0 OO

Angola 5 0 . 5 0 9 0

Uganda 7 - 3 . 2 7 oo

Kenya 6 4 1 . 3 2 2 7

Tanzania 5 0 . 2 4 3 9 8

Madagascar 9 0 . 0 0 OO

110 A D A N S O N I A , sér. 3 • 1997 • 1 9 ( 1 )

Flor is t ic r ichness in Af r i c a -Madagasca r

4 0 , were already on deposit at the t ime in the herbarium of the Laboratoire de Phanérogamie, Paris (P) alone.

Between 1981 and today, intensive field work has substantially increased the density index in a number o f countries, including Gabon , Ethiopia,

a n d M a d a g a s c a r . We cur ren t ly e s t i m a t e , for example, that as o f the end o f 1996 approximate­ly 5 5 , 5 0 0 c o l l e c t i o n s h a v e b e e n m a d e in Madagasca r since 1 9 8 1 , increasing the density index to at least 4 3 . If this level o f effort is main­tained, it will take about 51 years to reach a den-

10

1 A

§ 1 9 5 2 - 1 9 8 1 • 1974-1981

J1 i=n

CD

0) CO

o C3

I 4H

1952-1981 • 1974-1981

rr <

j f ] EL

TO 2

Fig 4 . — Collection rate in the Africa-Madagascar region (from CAMPBELL & HAMMOND 1989).

ADANSONIA, sér. 3 • 1997 • 19 (1 ) 111

M o r a t Ph. & L o w r y II P .P .

sity index o f 100 corresponding to a sufficient level o f floristic knowledge, contrary to what was suggested by C A M P B E L L & H A M M O N D (1989) . It

is nevertheless vividly clear that a t remendous effort is still required to generate an adequa te spec imen base , even for those for tunate areas where a reasonable level o f field work has been conducted; in the vast majority o f African coun­tries, however, the situation is regrettably much less encouraging.

In order to fill these t remendous gaps and to round out even our basic understanding o f floris­tic diversity in the Africa-Madagascar region, a number o f serious difficulties will first have to be overcome.

1. F inancial p rob lems o f varying m a g n i t u d e plague all o f the herbaria involved in the study o f the Afr ican flora, whether in E u r o p e , Africa, Nor th America, or elsewhere. Field work is chro­nically under-funded, staff levels are far too low, and current trends suggest that the situation will become even worse in the coming years.

2. Political instability, a lack o f security, and even the l ingering effects o f previous conflicts (e.g., the mill ions o f anti-personnel mines still buried in several African countries) hinder field exploration, and even effect the ability o f bota­nists to work in certain herbaria. As reported by Mats T H U L I N in 1993 (pers. comm.) , the herba­r i u m in M o g a d i s h u , S o m a l i a , was des t royed during the recent conflict in that country, and all the local botanists were in exile. T h e herbarium in Brazzaville, C o n g o , was seriously vandalized in early 1994, resulting in the loss o f all documents and equipment , and all unmounted specimens, although about 8 0 % o f the mounted collections were fortunately saved.

3. T h e number o f qualified taxonomists in pro­fessional positions that enable them to contribu­te to the study the flora o f the Africa-Madagascar region is steadily declining. Many projects have been or will soon be suspended because o f a lack o f pe r sonne l . F u r t h e r m o r e , several large a n d complex taxonomic groups (e.g., Rubiaceae and Euphorbiaceae) are nearly always left untreated, or are the last to be studied in ongoing flora pro­jects, largely because there are not enough spe­cia l is ts w o r k i n g on t h e m . T h i s de f i c iency is responsible in part for the slow progress o f cer­

tain flora pro jec ts , which were begun several decades ago, but remain unfinished. In the case of the Flore de Madagascar et des Comores , for example, while some of largest families are finally be ing treated (e .g . , Euphorb iaceae , Fabaceae) , virtually all o f the earlier treatments publ ished prior to 1960 mus t now be regarded as badly outdated and are in need o f complete revision, especially since the specimen base has grown so much in the last several decades.

As if the s i tuat ion presented above were not troublesome enough, we have not even consid­ered the status o f floristic investigation for the many groups o f cryptogams, about which even less is known. N o r have we touched on such cri­tically important disciplines as pollination biolo­gy, phenology, dispersal , growth, germinat ion , etc., which receive virtually no attention throu­g h o u t h u g e pa r t s o f the A f r i c a - M a d a g a s c a r region.

P R O S P E C T S F O R T H E F U T U R E

Despite the fact that our knowledge o f the flora o f the Africa-Madagascar region is far from com­plete (even though an average o f one species has been described per day since 1953 , according to L E B R U N 1991) , the available information base for the region is still more advanced than for either tropical Amer i ca or As ia . For this reason, the accumulated data on vascular plants from Africa and Madagascar, which now cover about 7 5 % of the species, are perhaps the best suited o f all for deve lop ing a comprehens ive project to verify, c o m p a r e a n d s t a n d a r d i z e i n f o r m a t i o n u s i n g modern computer technologies. Such an effort, if properly designed and implemented in a collabo­rative manner to take full advantage o f the avail­a b l e e x p e r t i s e , c o u l d r e a l i s t i c a l l y p r e p a r e a reliable, on-line computerized accounting o f the vascular plants o f the Africa-Madagascar region in as little as a decade.

Several projects deal ing with specific parts o f the region are now well under way, including the Conspectus o f the Vascular Plants o f Madagascar ( S C H A T Z et al. 1 9 9 6 ; P O N C Y & L A B A T 1 9 9 5 ) ,

which could serve as a very useful model for a larger undertaking. Similarly, the P R E C I S data-

112 ADANSONIA, ser. 3 • 1997 • 19 (1 )

Florist ic r ichness in Af r i ca -Madagasca r

base conta ins m u c h o f the base l ine botan ica l

i n fo rma t ion for S o u t h e r n Afr ica , a n d several

a d d i t i o n a l p r o j e c t s have l i k e w i s e s t a r t e d to

assemble relevant data from other areas, inclu­

ding L E A P (List o f East African Plants).

A computer ized project that treats the entire

African flora cou ld serve as an impor tan t and

useful precursor to a future Flora Africa, a syn­

thetic work first proposed by L E O N A R D in 1975 ,

but which has thus far not progressed beyond the

p l a n n i n g s t a g e . A c o m p u t e r i z e d d a t a b a s e o f

African plants would also facilitate better access

to a whole range o f valuable floristic informa­

tion, and thereby make it possible to assess more

accurately how well our knowledge of the flora o f

the Africa-Madagascar region is progressing. An

international effort, involving a broad range o f

institutions with complementary strengths, wor­

king together to study this important region, is

critically essential, and should be regarded as a

top priority for future botanical work in Africa.

A c k n o w l e d g m e n t s This paper was expanded and updated from a talk

entitled "Do we really know as much as we think about the flora of the Africa-Madagascar region?" pre­sented at the symposium on "Floras of the world: cur­rent knowledge and future prospects", organized by N . R . M O R I N and held at the X V International Botanical Congress, Yokohama, Japan, 3 Seprember 1993. We are grateful to the following persons for valuable information from work in progress on the African flora and/or from projects for which they are responsible: L. A N D R I A M I H E F A R I V O , F. B A D R E , P. B A M P S , J . B O S S E R , S. C A S T R O V I E J O , M. C H A L O P I N , C .

C U S S E T , J . - J . F L O R E T , F. F R I E D M A N N , L. G A U T I E R , P.

G O L D B L A T T , W. G R E U T E R , I. H E D B E R G , J . - P .

L E B R U N , I. M O R E I R A , R . P O L H I L L , G . P O P E , J .

R A H A R I M A M P I O N O N A , F. R A K O T O N D R A I N I B E , K.

S I K E S , G . F . S M I T H and M. T H U L I N .

REFERENCES

Anon. 1969.—Distributiones Plantarum Africanarum. Jardin Bot. Belg., Meise.

A K P A G A N A K. & B O U C H E T Ph. 1994.—État actuel des connaissances sur la flore et la végétation du Togo. Acta Bot. Gallica 141: 367-372.

A K P A G A N A K. & G U E L L Y K .A . 1 9 9 4 . — E s p è c e s d'Angiospermes nouvelles pour la Flore du Togo. Acta Bot. Gallica 141: 741-787.

A N D R E W F.W. 1950-1969.—Flowering plants of the Anglo-Egyptian Sudan, 3 vol. Buncle, Arbroath.

A U D R U J . , C É S A R J . & L E B R U N J . -P . 1994.—Les plantes vasculaires de la République du Djibouti. Vol. 1, 2 (in 2 parts), CSIRAD-EMVT.

BAMPS P. 1994.—Répertoire des familles de Phanérogames traitées dans les principales flores de LAfrique tropicale, Ed. 4. Jardin Botanique National de Belgique, Meise.

B E E N T J E H J . , A D A M S B . & D A V I S S . D . 1994 .— Regional overview: Africa: 101-148, in D A V I S S.D., H E Y W O O D V.H. & H A M I L T O N A.C. (eds.), Centres of Plant Diversity: A Guide and Strategy for their Conservation. Vol. 1, Europe, Africa, South West Asia and the Middle East. I U C N Puhl. Unit, Cambridge.

B O U L O S L. 1995.—Flora of Egypt checklist. El Hadara, Cairo.

B R E N A N J .P.M. 1978.—Some aspects of the phyto-geography of tropical Africa. ^4«». Missouri Bot. Gard. 65: 437-478.

B R U N E L J . F . , H I E P K O P. & S C H O L Z H . ( eds . ) . 1984.—Flore analytique du Togo. Phanérogames. Berlin & Eschborn, Englera: 1-751.

C A M P B E L L D.G. & H A M M O N D H.D. (eds.) 1989.— Floristic inventory of tropical countries. New York Bot. Gard., New York.

C O W L I N G R .M. , G I B B S R U S S E L L G .E . , H O F F M A N

M.T. & H I L T O N - T A Y L O R C. 1989.—Patterns of plant species diversity in southern Africa: 19-50, in H U N T L E Y B . J . (ed.), Biotic diversity in southern Africa: concepts and conservation. Oxford University Press, Cape Town.

E N G L E R A. & D l E L S L . 1936.—Syl labus der Pflanzenfamilien. 7 Aufl., Berlin.

F R I E D M A N N F. 1994 .—Flore des Seychelles. Dicotylédones. Vol. 1, O R S T O M , Paris.

F R O D I N D. G. 1984.—Guide to standard floras of the world. Cambridge University Press.

G O O D R. 197'4.— The geography of the flowering plants. 4th ed., London.

G R E E N G . M . & S U S S M A N R .W. 1 9 9 0 . — Deforestation history of the eastern rain forests of Madagascar from satellite images. Science 248: 212-215.

G R E U T E R W., B Ü R D E T H.M. & L O N G G. 1984.— Med-Checklist. Vol. 1, Cons. Jard. Bot. Genève.

G R E U T E R W., B Ü R D E T H.M. & L O N G G. 1986.— Med-Checklist. Vol. 3, Cons. Jard. Bot. Genève.

G R E U T E R W., B Ü R D E T H.M. & L O N G G. 1989.— Med-Checklist. Vol. 4. 458 p. Cons. Jard. Bot. Genève.

G U I N E A E. 1946.—Ensayo Geobotanico do Guinea continenta espagnola. Part 4, Flora, Madrid.

H E N D E R S O N A., C H U R C H I L L S.P. & L U T E Y N J . L .

1991.—Neotropical plant diversity. Nature 351 : 21-22.

H E P P E R 1979.—Second edition of the map showing the extent of floristic exploration in Africa South of the Sahara published by A E T F A T : 1 5 7 - 1 6 2 , in K U N K E L G. (ed.), Taxonomic aspects of African eco-

ADANSONIA, sér. 3 • 1997 • 19(1) 113

M o r a l Ph. & L o w r y II P.P.

nomic botany. Proc. IX Plenary Meeting of AET-FAT 1978.

H U M B E R T H . 1955.—Les territoires phytogéogra-phiques de Madagascar. Ann. BioL, sér. 3, 31: 439-448.

H U M B E R T H . 1959 (1960).—Origines présumées et affinités de la flore de Madagascar. Mém. Inst. Sei., Madag., sér. B, Biol. Vég., 9: 149-187.

K E M P , H . S . 1983.—A flora check-list for Swaziland. Occas. Papers 2, Swaziland Natl. Trust. Coram., Lobamba.

K O E C H L I N J . , G U I L L A U M E T J . - L . & M O R A T Ph.

1974.—Flore et végétation de Madagascar. J . Cramer Verlag, Vaduz.

L E B R U N J.-P. 1976.—Richesses spécifiques de la flore vasculaire de divers pays ou régions d'Afrique. Candollea 31 : 11-15.

L E B R U N J.-P. 1987.—Introduction à la connaissance de la Flore dAfrique. IEMVT, Maisons Alfort.

L E B R U N J.-P. 1997a.—Catalogue des plantes vasculaires de Mauritanie et du Sahara Occidental. C I R A D -EMVT, Montpellier (sous presse).

L E B R U N J.-P. 1997b.—Données nouvelles sur la flore de Djibouti. Candollea (sous ptesse).

L E B R U N J.-P., A U D R U J . , G A S T O N A. & M O S N I E R M. 1972.—Catalogue des plantes vasculaires du Tchad méridional. IEMVT, Maisons Alfort.

L E B R U N J . -P . & S T O R K A .L . 1977.—Index 1935-1976 des cartes de répartition des plantes vasculaires dAfrique. Cons. Jarcf. Bot. Genève.

L E B R U N J . -P . & S T O R K A .L . 1981.—Index 1935-1976 des cartes de répartition des plantes vasculaires dAfrique. Complément 1935-1976. Suppl. 1977-1981 avec addenda A - Z . Étude Bot . 8: 1-98. IEMVT, Maisons Alfort.

L E B R U N J.-P. & S T O R K A.L. 1991.—Enumeration des plantes à fleurs dAfrique tropicale. I. Cons. Jard. Bot., Genève.

L E B R U N J.-P. & S T O R K A.L. 1992.—Enumeration des plantes à fleurs dAfrique tropicale. II. Chryso-balanaceae à Apiaceae. Cons. Jard. Bot., Genève.

L E B R U N J.-P. & S T O R K A L . 1995.—Enumeration des plantes à fleurs d'Afrique tropicale. I I I . Monocotylédones: Limnocharitaceae à Poaceae. Cons. Jard. Bot., Genève.

L E B R U N J.-P. & S T O R K A . L . 1997.—Enumeration des plantes à fleurs dAfrique tropicale. IV. Gamopétales. Cons. Jard. Bot., Genève.

L E B R U N J.-P., T O U T A I N B „ G A S T O N A. & B O U D E T G. 1991.—Catalogue des plantes vasculaires du Burkina Faso. Études et Synthèses de l'IEMVr, no. 40.

L É O N A R D J . 1968.—Statistiques des progrès accomplis en 13 ans dans la connaissance de la flore phanéroga-mique africaine et malgache (1953-1965): 297-299, in H E D B E R G O. (ed.), Conservation of végétation in Africa South ofthe Sahara. Proc. 6th Meeting AET-FAT, Acta Phytogeogi. Suecica, vol. 54.

L É O N A R D J . 1994.—Statistiques des spermatophytes de la Flore d'Afrique Centrale de 1940 à 1990.

Bull. fard. Bot. Nat. Belg. 63: 181-194. L O W R Y P. P. II, S C H A T Z G .E . & P H I L L I P S O N P.B.

1997.—The classification of natural and anthropo­genic vegeta t ion in Madagasca r : 9 3 - 1 2 3 , in G O O D M A N S . M . & P A T T E R S O N B . D . ( eds . ) , Natural change and human impact in Madagascar. Smithsonian Inst. Press, Washington, D.C.

M A I R E R. 1952-1987.—Flore de lAfrique du Nord. Maroc, Algérie, Tunisie, Tripolitaine, Cyrénaïque et Sahara. 16 vol., 73 fase, Lechevalier, Paris.

M E R X M Ü L L E R H. et al. 1966-1972.—Prodromus einer Flora von Südwestafrika. Bot. Mus., Berlin.

P H I L L I P S O N P. B . 1994.—Madagascar: 271-281, in D A V I S S.D., H E Y W O O D V.H. & H A M I L T O N A.C.

(eds.), Centres of plant diversity. A guide and strategy for their conservation. Vol. 1, Europe, Africa, South West Asia and the Middle East, l U C N Pubi. Unit, Cambridge.

P O N C Y O. & L A B A T J . - N . 1995.—État actuel de l'inventaire des flores tropicales. Biosystema 13 : 23-50.

PRANCE G. T. 1977.—Floristic inventory of the tro­pics: where do are stand? Ann. Missouri Bot. Gard. 64: 659-684.

QuÉZEL P. 1985.—Definition of the Mediterranean region and the origin of its flora: 17, in GoMEZ-C A M P O C . ( ed . ) , Plant Conservation in the Mediterranean area. Junk, The Hague.

S C H A T Z G.E., L O W R Y II P.P., L E S C O T M „ W O L F A.-

E., A N D R I A M B O L O L O N E R A S., R A H A R I M A L A L A V. &

R A H A R I M A M P I O N O N A J . 1996.—Conspectus of the vascular plants of Madagascar: a taxonomic and conservation electronic database: 10-17, in V A N D E R M A E S E N L.J.G. et al. (eds.), The Biodiversity of African Plants. Kluwer Acad. Pubi., Dordrecht, Boston and London.

S M I T H G . F . , D A V I S G .W. , D O N A L D S O N J . S . &

R O U R K E J .P . 1995.—Research in the National Botanical Institute of South Africa. S. Afr. ]. Sci. 91: 597-640.

SOUZA S. 1987.—Flore du Bénin. Vol. 1: Catalogue des plantes du Bénin. Vol. 2: Plantes du bord de mer et de mangrove. Mission d'aide et de coopéra­tion et Université Nationale du Bénin, Cotonou.

S T O R K A. L. & L E B R U N J . -P. 1988.—Index 1935-1976 des cartes de répartition des plantes vasculaires dAfrique. Complément 1935-1981. Suppl. 1982-1985. IEMVT, Maisons Alfort.

TAKHTAJAN A. 1986.—Floristic Regions ofthe World. University of California Press, Berkeley.

T U R I L L W.B. 1959.—Plant Geography: 172-229, in C O L L I N S & T A Y L O R G. (eds.), Vistas in Botany, vol. 2, London.

VlNK W. 1 9 8 1 . — D e n s i t y indexes updated . Fl. Malesiana Bull. 34: 3567-3568.

W H I T E F. 1983.— The vegetation of Africa. A descrip­tive memoir to accompany the UNESCO!AETFATI UNSO Vegetation map of Africa. U N E S C O , Paris.

W H I T E F. 1993.—The AETF A T chorological classifi-

114 ADANSONIA, s é r . 3 • 1 9 9 7 • 1 9 ( 1 )

Flor is t ic r ichness in Af r i c a -Madagasca r

cation of Africa: history, methods and applications. BulLJard. Bot. Nat. Belg. 6 2 : 2 5 5 - 2 8 1 .

World Conservation Monitoring Centre 1 9 8 2 . — Global biodiversity status of Earth's living resources. Chapman & Hall, London.

Flora Capensis ( 1 8 6 0 - 1 9 3 3 ) . — H A R V E Y & S O N D E R (eds.), London.

Flora of Egypt ( 1 9 4 1 - 1 9 6 9 ) . — T A C K H O L M V . & G., 4 vol., Cairo.

Flora of Ethiopia ( 1 9 8 9 - ) . — H E D B E R G I. ( ed . ) , Uppsala.

Student's Flora of Egypt 1 9 7 4 . — T A C K H O L M V . , ed. 2 , 1 vol., Cairo.

Flora of Somalia ( 1 9 9 3 - ) . — T H U L I N M. (ed.), Royal Bot. Gard., Kew.

Flora of Southern Africa ( 1 9 6 3 - ) . — D Y E R R.A. &

C O D D L .E. (eds.), Departement of Agriculture, Technical Services, Pretoria; from 1 9 7 9 L E I S T N E R O.A. (ed.), Pretoria.

Flora of West Tropical Africa ( 1 9 5 4 - 1 9 7 2 ) . — H U T C H I N S O N J . & D A L Z I E L J . M . , ed. 2 , 2 vol., Millbank, London.

Flore de l'Afrique du Nord (1952-1987) .—Lecheval ier , 1 6 vol., Paris.

Flore du Cameroun ( 1 9 6 3 - ) . — M u s . Natl. Hist. Nat., Laboratoire de Phanérogamie (ed.), Paris; since 1 9 8 0 , with MSIRES, Yaounde.

Flore du Congo belge et du Ruanda-Urundi ( 1 9 4 8 - ) . — INEAC, Bruxelles; since 1 9 7 5 , Flore de l'Afrique Centrale, Jardin Bot. National de Belgique, Bruxelles.

Flore du Gabon ( 1 9 6 1 - ) . — M u s . Natl . Hist. Nat. , Laboratoire de Phanérogamie (ed.), Paris.

Flore des Mascareignes ( 1 9 7 6 - ) . — M S I R I , O R S T O M & Kew, Port Louis and Paris.

Manuscript received 9 September 1996; revised version accepted 17 February 1997.

ADANSONIA, sér. 3 • 1997 • 19 (1 ) 115