Search

Tree size shapes forest carbon dynamics and determines how trees interact with their environment, including a changing climate. Here, we conduct the first global analysis of among-site differences in how aboveground biomass stocks and fluxes are distributed with tree size. We analyzed repeat tree censuses from 25 large-scale (4-52 ha) forest plots spanning a broad climatic range over five continents to characterize how aboveground biomass, woody productivity, and woody mortality vary with tree diameter. We examined how the median, dispersion, and skewness of these size-related distributions vary with mean annual temperature and precipitation. In warmer forests, aboveground biomass, woody productivity, and woody mortality were more broadly distributed with respect to tree size. In warmer and wetter forests, aboveground biomass and woody productivity were more right skewed, with a long tail towards large trees. Small trees (1-10 cm diameter) contributed more to productivity and mortality than to biomass, highlighting the importance of including these trees in analyses of forest dynamics. Our findings provide an improved characterization of climate-driven forest differences in the size structure of aboveground biomass and dynamics of that biomass, as well as refined benchmarks for capturing climate influences in vegetation demographic models.

1. Organisms of all species must balance their allocation to growth, survival and recruitment. Among tree species, evolution has resulted in different life-history strategies for partitioning resources to these key demographic processes. Life-history strategies in tropical forests have often been shown to align along a trade-off between fast growth and high survival, that is, the well-known fast–slow continuum. In addition, an orthogonal trade-off has been proposed between tall stature—resulting from fast growth and high survival—and recruitment success, that is, a stature−recruitment trade-off. However, it is not clear whether these two independent dimensions of life-history variation structure tropical forests worldwide. 2. We used data from 13 large-scale and long-term tropical forest monitoring plots in three continents to explore the principal trade-offs in annual growth, survival and recruitment as well as tree stature. These forests included relatively undisturbed forests as well as typhoon-disturbed forests. Life-history variation in 12 forests was structured by two orthogonal trade-offs, the growth−survival trade-off and the stature−recruitment trade-off. Pairwise Procrustes analysis revealed a high similarity of demographic relationships among forests. The small deviations were related to differences between African and Asian plots. 3. Synthesis. The fast–slow continuum and tree stature are two independent dimensions structuring many, but not all tropical tree communities. Our discovery of the consistency of demographic trade-offs and life-history strategies across different forest types from three continents substantially improves our ability to predict tropical forest dynamics worldwide.

The high mortality and low recruitment of the myrmecophytic Acacia drepanolobium are not consistent with the demographic rates of monodominant species. The positive conspecific density dependence observed from the spatial analysis is consistent with the defensive benefits conferred by symbiotic ants to trees when they grow close to each other. Patches of savanna dominated by Acacia drepanolobium occur throughout East Africa on nutrient-rich vertisols, also known as black cotton soils. We assessed the survival and recruitment for all freestanding trees with diameter at knee height (dkh) ≥ 10 mm in one of such mono-dominated patches (47 ha) at the Mpala Research Centre, Kenya, with the aim of identifying demographic traits that might explain the dominance of this species. Over a mean 6-year interval, mortality and recruitment rates in the habitat were 4.55%/year and 1.42%/year respectively, resulting in a net loss of 17.8% of the initial individuals. Of the 30 species recorded from the first census, 11 decreased in abundance, nine increased, and the remainder 10 did not change in abundance. The monodominant A. drepanolobium had a high mortality (4.69%/year), a low recruitment (1.31%/year), and a 19% population decline. There was no evidence of conspecific negative density dependence for this species. Rather, we found a statistically significant positive correlation between the number of conspecific neighbors and individual-level probability of survival, consistent with the “shared defense” benefits that symbiotic ant colonies occupying multiple trees can confer to these latter in a small neighborhood. Thus, mortality of A. drepanolobium was higher in areas where it occurred in lower densities, which resulted in an increase in the spatial aggregation of conspecifics. Mortality increased with dkh size classes and was mostly caused by elephants and stem-boring beetles. The demographic rates during the study period in theory are inconsistent with those of monodominant species. The protection against herbivory conferred by mutualistic ants associated with this species remains the most probable explanation of its dominance in this habitat.

QUESTIONS: We examined the spatial patterns of dominant canopy species in a tropical forest to investigate: (a) what is the niche occupancy of canopy species with respect to topographic gradients; (b) what are the dominant ecological processes that explain their distribution; (c) what are the interactions among the most prevalent canopy species; and (d) what are the interactions between canopy species adults and juveniles trees? LOCATION: Rabi permanent CTFS‐ForestGEO plot, Gabon. METHODS: We selected the four most abundant canopy species and one timber species. We used Berman's test to determine the effect of three topographic variables on the distribution of each species and univariate analysis to model the spatial pattern of each species using either an inhomogeneous Poisson process or an inhomogeneous Cox process. We also used a bivariate form of the pair correlation function (PCF) to determine the spatial interaction between species and the correlation among conspecific adult and juvenile trees. RESULTS: Four of the five species had aggregated spatial patterns while Lophira alata showed spatial randomness. Most of the variance in the local tree density was explained by within‐population dispersal processes rather than environmental factors. Bivariate PCF tests showed significant segregation between species associations. Two species exhibited aggregation at small distances between young and adult trees, while others showed either complete spatial randomness at small inter‐tree distances or segregation at large distances. CONCLUSIONS: This study showed that the spatial pattern in the majority of canopy species was aggregation. Seed dispersal limitation mainly explained the observed aggregation pattern. Habitat filtering, as evidenced by the influence of topographic variables on niche occupancy, marginally, yet significantly, explained this pattern. The different spatial patterns of the principal species permit their coexistence. Spatial segregation among adult and juvenile trees reveals a strong pattern of either species‐specific seed predation or pathogens.

Background and aims – Cola, the second largest genus of the Malvaceae-Sterculioideae comprises 100–135 small to large tree species confined in nature to African forests, though cultivated elsewhere. Current species distribution ranges show that the genus is highly diverse in the seasonally wet forests along the Nigeria-Cameroon border, including the Korup National Park (KNP). In this paper we examine the diversity and abundance of Cola in KNP compared to other forests for which comparable data are available. We also describe two novelties in the genus.Methods – We used inventory data from a 50-ha permanent plot in southern KNP where all Cola trees and saplings down to 1 cm in diameter were tagged, mapped and identified. Additional collections of the genus came from the 11 km trail leading to the plot. Classic herbarium techniques and field observations were used for the morphological identification and description of specimens at MO and YA and from our personal collections. Cola species richness and abundance was estimated from the plot data and compared to other African forest sites for which comparable data are available. The evaluation of the conservation status of the two new species described in this paper followed the IUCN Red List Categories and Criteria. Key results – Twenty-five species of Cola were identified in the southern part of the KNP, including four undescribed species, raising the total number of Cola species in Cameroon to 46. The abundance of the genus in KNP was three orders of magnitude higher than in the Rabi forest in southwestern Gabon or in the Ituri forest in eastern D.R. Congo. This high species richness and abundance suggests that KNP is part of the center of diversity of the genus. Two new species, Cola zemagoana Kenfack & D.W.Thomas and C. mamboana Kenfack & Sainge are described and illustrated. Both species are only known from the lowland rainforest of southwestern Cameroon. Cola zemagoana is narrow endemic of southern KNP and its conservation status is assessed as Endangered. Cola mamboana is confined to the lowland forests of southwestern Cameroon, is locally very abundant in protected areas and is also assigned the conservation status Endangered.

Rhaptopetalumrabiense Kenfack & Nguema, sp. nov. from the Rabi forest in south-western Gabon is described, illustrated and assigned a provisional conservation status of “Critically Endangered”. An identification key to the five Gabonese species of Rhaptopetalum is also provided.

Background: The Rumpi Hills Forest Reserve (RHFR) is a montane forest area in south-western Cameroon. Although RHFR is presumed to be rich in biodiversity and vegetation types, little information exists regarding its floristic composition and vegetation patterns. Objectives: Our goal was to characterise vegetation patterns in the reserve and to understand how elevation influences distributions and diversity of species. We aimed to provide a first detailed plant species inventory for this important forest area, as well as basic information on forest structure. Method: We characterised floristic composition and vegetation patterns of the reserve in 25 1-ha plots along an elevational gradient from 50 m to 1778 m. In each plot, trees and lianas of diameter at breast height (dbh) ≥ 10 cm were measured; shrubs < 10 cm were measured in nested plots of 0.01 ha. Results: In all, 16 761 trees, shrubs and lianas with dbh ≥ 1 cm were censused, representing 71 families, 279 genera and 617 morphospecies. Floristic composition ranged from 94 to 132 species, with a mean of 117.5 species per hectare in lowland forest (50 m – 200 m) and 36–41 species, with a mean of 38.5 species per hectare in montane cloud forest (1600 m – 1778 m) near the summit of Mount Rata. Two-way indicator species analysis classified the 25 plots into six vegetation types corresponding to lowland evergreen rainforest, lowland evergreen rainforest on basalt rocks, middle-elevation evergreen forest, submontane forest, transitional submontane forest and montane cloud forest. In all, 0.006% of the reserve was included in our sample plots. Detrended correspondence analysis highlighted the importance of elevation in shaping vegetation patterns. Conclusion: The RHFR is composed of different vegetation types, which show impressive variation in terms of structure, species composition and diversity. The detailed, fine-scale inventory data obtained in this study could be useful in planning efficient management of this and other montane tropical forests.

Gautier, L., O. Lachenaud, X. Van der Burgt & D. Kenfack (2016). Five new species of Englerophytum K. Krause (Sapotaceae) from central Africa. Candollea 71 ≥ : 287–305. In English, English and French abstracts. Five new species of Englerophytum K. Krause (Sapotaceae) are described : Englerophytum paludosum L. Gaut., Burgt & O. Lachenaud, Englerophytum gigantifolium O. Lachenaud & L. Gaut., Englerophytum libenii O. Lachenaud & L. Gaut., Englerophytum sylverianum Kenfack & L. Gaut., and Englerophytum ferrugineum L. Gaut. & O. Lachenaud. All five species are illustrated with line drawings and three of them with field photos. Distribution maps are also provided, and a preliminary extinction risk assessment according to IUCN Categories and Criteria is carried out : Englerophytum paludosum is assessed as “Least Concern”, Englerophytum sylverianum as “Vulnerable”, Englerophytum libenii and Englerophytum ferrugineum as “Endangered”, and Englerophytum gigantifolium as “Critically Endangered”.

The African genus Crotonogynopsis Pax (Euphorbiaceae) is revised to include four species, including two novelties, C. korupensis Kenfack & D. W. Thomas from the Korup National Park, Cameroon, and the Reserva Natural de Rio Campo in Equatorial Guinea; and C. australis Kenfack & Gereau from the southern part of the Eastern Arc Mountains of Tanzania, with a distant outlier in Mozambique. Three of the four species are assigned the IUCN Red List category of Least Concern (LC) on the basis of their occurrence in protected areas with no known threats.

Background and aims – There is increasing recognition that plant traits mediate environmental influence on species distribution, justifying non-random community assembly. We studied the influence of local scale edaphic factors on the distribution of functional traits in a tropical rainforest of Cameroon with the aim to find correlations between the main edaphic gradient and community functional trait metrics (weighted mean trait, functional divergence and intraspecific variation). Methods – Within the Korup Forest Dynamics Plot (50 ha), we randomly selected 44 quadrats of 0.04 ha each, collected soils and analysed 11 topography and soil variables. Leaves were harvested from all 98 tree species found in the quadrats to calculate community trait metrics [quadrat-level weighted mean (qk) and functional divergence (FDivk)] for leaf area (LA), specific leaf area (SLA), leaf phosphorus (LPC), leaf nitrogen concentration (LNC) and nitrogen to phosphorus ratio (N:P ratio). We examined relationships between the main edaphic gradient with qk, with FDivk and with intraspecific variation and interpreted correlations as the effects of abiotic filtering and competitive interaction. Key results – Soil fertility was the main edaphic gradient and was significantly correlated with qk for LPC, LNC and LA and with FDivk for LPC, N:P ratio, LA and SLA, confirming the influence of abiotic filtering and competitive interaction by the soil fertility gradient, respectively. For a given trait, quadrats were either over-dispersed or under-dispersed, accounting for 7–33 % of non-random trait distribution along the soil fertility gradient. Trends in intraspecific traits variation were consistently lower than quadrat-level mean traits along the soil fertility gradient. Conclusions – This study demonstrates the influence of soil fertility gradient on local scale community trait distribution and its contribution to non-random community assembly., SCOPUS: ar.j, info:eu-repo/semantics/published

Ant-plant interactions mediated by special structures provided by plants such as domatia, extrafloral nectaries (EFNs) and food bodies, are very frequent in tropical ecosystems. To understand why ants are frequently encountered on most species of Carapa Aubl. (Meliaceae), we investigated the presence of extranuptial nectaries (ENNs) in all 27 species of the genus, spanning its entire distributional range in tropical Africa and America. We report for the first time in the genus the occurrence of extrafloral nectaries (at the base of the petiole, along the rachis of the pinnately compound leaf, on bracts) petaline nectaries (on the outer surface of petals), and pericarpial nectaries (on the surface of fruits), and confirm the presence of nectaries on leaflets in Carapa. Petiolar nectaries are the most common, occurring in 85% of the species. Nectaries were mainly active in young developing plant organs. Ants were observed foraging on exudates from these nectaries. The secretions from these glands help to explain the abundance of ants on Carapa trees. Although similar nectaries were also found in other members of the subfamily Cedreloideae, their position and frequency provide new characters for the identification of Carapa species in the field and the herbarium. As in other myrmecophilous plants, ENNs probably confer adaptive advantages to Carapa trees.

Morphometric analysis of Phyllobotryon Muell.Arg. in the Korup Forest Area of Cameroon recognizes three distinct morphospecies (1-3), which show significant variation in several leaf and fruit characters. In order to clarify the taxonomy of Phyllobotryon, we conducted univariate and multivariate analyses on sixteen quantitative and four qualitative characters scored from 111 fresh samples. Analysis of Variance revealed nine significant quantitative characters from which the first three Principal Components accounted for 74.6 % of the total variation. Results from Discriminant Analysis strongly support the existence of two groups (96.2 % and 100 %) representing Morphospecies 2 and 3, but morphospecies 2 is only weakly supported (88.9 %) as distinct from morphospecies 1. Characters such as petiole length, fruit surface ornamentation, style & calyx persistence, flower and fruit orientation and leaf shape are of

Summary 1. In highly diverse ecosystems, such as tropical forests, the relative importance of mechanisms underlying species coexistence (e.g. habitat filtering, competitive exclusion, neutral dynamics) is still poorly known and probably varies depending on spatial and phylogenetic scales. 2. Here, we develop new approaches for dissecting simultaneously the phylogenetic structure of communities at different phylogenetic depths and spatial scales. We tested with simulations that our method is able to disentangle overdispersion and clustering effects occurring at contrasted phylogenetic depths. 3. We applied our approaches to a 50 ha Forest Dynamic Plot located in Korup National Park (Cameroon) where 329,000 tree stems ≥ 1 cm in diameter were identified and mapped, and using a newly generated dated molecular phylogenetic tree based on 2 plastid loci (rbcL and matK), including 272 species from Korup (97% of the individuals). 4. Significant patterns of phylogenetic turnover were detected across 20 9 20 m 2 quadrats at most spatial scales, with higher turnover between topographic habitats than within habitats, indicating the prevalence of habitat filtering processes. Spatial phylogenetic clustering was detected over the entire range of phylogenetic depths indicating that competitive exclusion does not generate a pattern of phylogenetic overdispersion at this scale, even at a shallow phylogenetic depth. 5. Using an individual-based approach, we also show that closely related species tended to aggregate spatially until a scale of 1 m. However, the signal vanishes at smaller distance, suggesting that competitive exclusion can balance the impact of environmental filtering at a very fine spatial scale. 6. Synthesis. Using new methods to characterize the structure of communities across spatial and phylogenetic scales, we inferred the relative importance of the mechanisms underlying species coexistence in tropical forests. Our analysis confirms that environmental filtering processes are key in the structuring of natural communities at most spatial scales. Although negative-density tends to limit coexistence of closely related species at very short distance (

Summary. Two new species of Afrothismia are described within the framework of an intended revision of the genus in Cameroon, A. fungiformis from the submontane forest of Mt Kupe and A. pusilla from Mt Kala. The new species are illustrated and the conservation status for both is assessed as Critically Endangered.

The relationship between species richness and ecosystem function, as measured by productivity or biomass, is of long-standing theoretical and practical interest in ecology. This is especially true for forests, which represent a majority of global biomass, productivity and biodiversity. Here, we conduct an analysis of relationships between tree species richness, biomass and productivity in 25 forest plots of area 8–50 ha from across the world. The data were collected using standardized protocols, obviating the need to correct for methodological differences that plague many studies on this topic. We found that at very small spatial grains (0.04 ha) species richness was generally positively related to productivity and biomass within plots, with a doubling of species richness corresponding to an average 48% increase in productivity and 53% increase in biomass. At larger spatial grains (0.25 ha, 1 ha), results were mixed, with negative relationships becoming more common. The results were qualitatively similar but much weaker when we controlled for stem density: at the 0.04 ha spatial grain, a doubling of species richness corresponded to a 5% increase in productivity and 7% increase in biomass. Productivity and biomass were themselves almost always positively related at all spatial grains. Synthesis. This is the first cross-site study of the effect of tree species richness on forest biomass and productivity that systematically varies spatial grain within a controlled methodology. The scale-dependent results are consistent with theoretical models in which sampling effects and niche complementarity dominate at small scales, while environmental gradients drive patterns at large scales. Our study shows that the relationship of tree species richness with biomass and productivity changes qualitatively when moving from scales typical of forest surveys (0.04 ha) to slightly larger scales (0.25 and 1 ha). This needs to be recognized in forest conservation policy and management.

Gambeya korupensis Ewango & Kenfack (Sapotaceae: Chrysophylloideae), a new rain forest tree species from the Southwest Region in Cameroon, is described and illustrated. A distribution map is provided. G. korupensis has the leaf blade below pubescent on the midribs and secondary nerves, flowers with a pedicel 0.5 – 1 mm long, and a fruit which is ovoid, attenuate at the apex, 5-ridged, verrucose between the ridges, and bright red at maturity. The conservation status of G. korupensis is assessed as Vulnerable according to IUCN criteria.

Tropical forests have long been recognized for their biodiversity and ecosystem services. Despite their importance, tropical forests, and particularly those of central Africa, remain understudied. Until recently, most forest inventories in Central Africa have focused on trees ≥10 cm in diameter, even though several studies have shown that small-diameter tree population may be important to demographic rates and nutrient cycling. To determine the ecological importance of small-diameter trees in central African forests, we used data from a 25-ha permanent plot that we established in the rainforest of Gabon to study the diversity and dynamics of these forests. Within the plot, we censused 175,830 trees ≥1 cm dbh from 54 families, 192 genera, and 345 species. Average tree density was 7,026 trees/ha, basal area 31.64 m2/ha, and above-ground biomass 369.40 Mg/ha. Fabaceae, Ebenaceae and Euphorbiaceae were the most important families by basal area, density and above-ground biomass. Small-diameter trees (1 cm ≥ dbh

A new species of Cassipourea is described from the banks of the Mana River, in the southeastern boundary of the Korup National Park in Cameroon and placed provisionally within the subgenus Cassipourea. The leaves of the new species are close to those of C. afzelii, but its flowers differ from those of the latter and the rest of the species of the subgenus Cassipourea in having a glabrous ovary and a diplostemonous androceum, with filaments distinctly of two lengths. The illustration of the new species is provided and based on its narrow distribution, the provisional conservation status Vulnerable VU D1 is assigned.

Carapa comprises trees confined to tropical forests in Africa and America. Previous revisions of the genus based primarily on herbarium specimens recognized few variable species, but also stressed the need for further research to clarify and understand the patterns of morphological variation in these species complexes. In the present revision, 27 species are recognized in Carapa of which 16 occur in Africa and 11 in America. Nine new species are described and illustrated, and a new combination is proposed. The remaining 17 species correspond to either recently described species or previously described species that had been placed in synonymy. A key to the 27 species and their distribution maps are provided.

Niche differentiation with respect to habitat has been hypothesized to shape patterns of diversity and species distributions in plant communities. African forests have been reported to be relatively less diverse compared to highly diversed regions of the Amazonian or Southeast Asian forests, and might be expected to have less niche differentiation. We examined patterns of structural and floristic differences among five topographically defined habitats for 494 species with stems ≥1 cm dbh in a 50-ha plot in Korup National Park, Cameroon. In addition, we tested for species–habitat associations for 272 species (with more than 50 individuals in the plot) using Torus translation randomization tests. Tree density and basal area were lowest in areas with negative convexity, which contained streams or were inundated during rainy periods and highest in moist well-drained habitats. Species composition and diversity varied along the topographical gradient from low flat to ridge top habitats. The low depression and low flat habitats were characterized by high diversity and similar species composition, relative to slopes, high gullies and ridge tops. Sixty-three percent of the species evaluated showed significant positive associations with at least one of the five habitat types. The majority of associations were with low depressions (75 species) and the fewest with ridge tops (8 species). The large number of species–habitat associations and the pronounced contrast between low (valley) and elevated (ridgetop) habitats in the Korup plot shows that niche differentiation with respect to edaphic variables (e.g., soil moisture, nutrients) contributes to local scale tree species distributions and to the maintenance of diversity in African forests.

Carapa vasquezii from western Amazonia is described and illustrated. The new species differs from the closely related C. guianensis in having leaflets with an obtuse to rounded apex and a farinose midrib, farinose inflorescence and flower pedicels, 6-ovulate locules, and smaller seeds with a very much reduced hilum.

A taxonomic revision of Carapa (Meliaceae) in Ecuador is provided with the recognition of four species, C. megistocarpa which has cauliflorous inflorescences, C. nicaraguensis previously described and currently placed as synonym of C. guianensis and two new species (C. alticola and C. longipetala). The new species are close to C. guianensis based on their 4-merous flowers borne at the end of the branches. However, C. alticola differs from C. guianensis in having larger leaflets with prominent secondary veins, seeds with rounded edges and short poorly ramified inflorescences, while C. longipetala can be distinguised from C. guianensis in having distinctly pedicellate flowers and 6-ovulate ovary loculi. The new species are described, illustrated, and a key to the four species recognized in Ecuador is provided.

The taxonomy of the amphi-Atlantic tree genus Carapa (Meliaceae) has long been controversial. Of the three species currently recognized in the genus, two are known to present substantial morphological variation that has been used in the past to distinguish several taxa, most of which are currently placed in synonymy. Here, a combination of field observations, univariate analyses of leaf, floral and seed characters and principal coordinate analyses of floral characters in the context of a molecular phylogenetic analysis was used to investigate the patterns of variation and delimit morphological species anew in the genus. These results support the recognition of 27 species in Carapa, of which 16 are previously described and 11 are new. In general, phylogenetically related species occurred in the same geographical area, but were morphologically distinct. © 2011 The Linnean Society of London, Botanical Journal of the Linnean Society, 2011, 165, 186–221.

A distinctive new species of Rubiaceae from Rio Muni, Equatorial Guinea, is described and illustrated. Leptactina rheophytica is the only rheophyte known in the genus. Its diagnostic characters are elucidated, its taxonomic affinities are discussed, and notes on its conservation status are provided. © 2007 The Linnean Society of London.

A new species of Triuridaceae, Kihansia jengiensis Sainge & Kenfack is described from the semi-deciduous forest of Southeastern Cameroon. The new species is illustrated and a key to the identification of the two species in the genus provided. The species constitutes the first record of the genus in central Africa and its conservation status is assessed as Critically Endangered.

The new species Cassipourea korupensis Kenfack & Sainge (Rhizophoraceae) from western Cameroon is described and illustrated, and its conservation status discussed. Unlike other species of subgenus Lasiosepalum Alston, in which it is provisionally placed, the new species has a distinctive glabrous ovary and a cupuliform calyx.

Carapa akuri, a new species endemic to central Guyana, is described and illustrated. It is compared to the two other species (C. guianensis and C. surinamensis) occurring in the Guianas.

Between 1990 and 2000 a number of vegetation surveys were undertaken on the Mount Cameroon area using mainly three methods: 0.25 ha sample plots, 20 m wide transects and rapid botanical surveys. A comparison of these methods in terms of data collected, operational characteristics and output relevance, revealed their strengths and weaknesses. The methods vary markedly in their ecological value (0.25 ha sample plots superior), taxonomic scope and rigour (rapid botanical survey superior) and logistic efficiency (transects superior). The combined outcome of all these surveys is discussed and recommendations made on the use of the rapid botanical survey in order to rapidly fill the remaining gaps on plant distribution.

During the past decade, two long-term studies of tropical forest trees and lianas were initiated in the Korup National Park, Cameroon and in the Reserve de Faune 'a Okapis, Democratic Republic of Congo. These two large forest dynamics programs are part of a global network of research sites coordinated by the Smithsonian Tropical Research Institute, and are leading to the discovery of many new plant species (e.g., Gereau & Kenfack 2000; Ewango & Breteler 2001; Sonk6 et al. 2002). During the study of the Korup plot vouchers at BR, MO, K, SCA and YA, some specimens belonging to the genus Manilkara Adans. could not be matched, and did not fit any of the species described from western and central Africa (Aubreville 1961, 1964; Gautier 1997). Critical examination of the Korup specimens revealed sufficient differences to justify the description of a new species. Further surveys in the Korup area to collect additional specimens of this taxon have so far been unsuccessful. This description is therefore based on a very small number of individuals, and we have little data on intra-specific variation. However, since we have both flowering and fruiting material showing a suite of distinctive characters, there is no reason to delay publication. The new species clearly belongs in the genus Manilkara because of the calyx, which has two whorls of 3 sepals each. Although Manilkara species are often difficult to separate, ours does not resemble any other described species very closely. M. pellegriniana Tisser. & Sillans is the only other species from the region with persistent stipules and glabrous, laciniate staminodes, and the new species can be distinguished from it by a number of well-defined characters. Manilkara lososiana Kenfack & Ewango sp. nov. Manilkarae pellegrinianae Tisser. & Sillans similis, sed stipulis triangularibus non filiformibus, petiolo breviore 5 15 non 20 40 mm longo, pedicello longiore, foliorum indumento brunneo (nec griseoargenteo) distinguitur. Typus: Cameroon, Southwest Province, Korup National Park, Kenfack 625 (holotypus YA; isotypi SCA, MO, K).

Uvariopsis submontana Kenfack, Gosline & Gereau (Annonaceae) is described from premontane habitats in western Cameroon. The new species is morphologically most similar to U. korupensis Gereau & Kenfack, recently described from lower elevations in western Cameroon. The monotypic genus Dennettia Baker f. is completely consistent with the morphological and palynological circumscription of Uvariopsis Engler, and is included in the latter genus under the new combination U. tripetala (Baker f.) G. E. Schatz. A key to the 16 species of this tropical African genus is presented. R1ASUM£. Uvariopsis submontana Kenfacke Gosline & Gereau (Annonaceae) est decrit d'habitats premontagnards de l'Ouest du Cameroun. Quant a sa morphologie, la nouvelle espece ressemble au plus pres 1'U. korupensis Gereau & Kenfack, recemment decrit d'altitudes plus basses de l'Ouest du Cameroun. Le genre monotypique Dennettia Baker f. est tout a fait consistent avec la circonscription morphologique et palynologique d'Uvariopsis Engler et est inclu dans celui-ci sous la nouvelle combinaison Uvariopsis tripetala (Baker f.) G. E. Schatz. Une cle des 16 especes de ce genre de l'Afrique tropicale est presentee.