Occurrence of nodulation in unexplored leguminous trees native to the West African tropical rainforest and inoculation response of native species useful in reforestation


  • Moussa Diabate,

    1. Institut de Recherche Agronomique de Guinée, Division des Cultures Pérennes, Programme Recherche Forestière, BP 1523, Conakry, République de Guinée;
    Search for more papers by this author
  • Antonio Munive,

    1. Laboratorio de Microbiologia del Suelo, Instituto de Ciencias, Benemerita Universidad Autonoma de Puebla, Mexico;
    Search for more papers by this author
  • Sérgio Miana De Faria,

    1. EMBRAPA, Agrobiologia, BR 465, Km 47, CEP 23.890-000, Seropédica, Rio de Janeiro, Brazil;
    Search for more papers by this author
  • Amadou Ba,

    1. Laboratoire de Biologie et Physiologie Végétales, Faculté des Sciences Exactes et Naturelles, Université des Antilles et de la Guyane, BP 592, 97159 Point-à-Pitre, Guadeloupe, France;
    Search for more papers by this author
  • Bernard Dreyfus,

    1. Laboratoire des Symbioses Tropicales et Méditerranéennes, UMR 113 (IRD/CIRAD/INRA/ENSAM), Campus International de Baillarguet, TA 10/J, 34398 Montpellier Cedex 5, France
    Search for more papers by this author
  • Antoine Galiana

    Corresponding author
    1. Laboratoire des Symbioses Tropicales et Méditerranéennes, UMR 113 (IRD/CIRAD/INRA/ENSAM), Campus International de Baillarguet, TA 10/J, 34398 Montpellier Cedex 5, France
      Author for correspondence: Antoine Galiana Tel: +33 4 67 59 38 51 Fax: +33 4 67 59 38 02 Email: galiana@cirad.fr
    Search for more papers by this author

Author for correspondence: Antoine Galiana Tel: +33 4 67 59 38 51 Fax: +33 4 67 59 38 02 Email: galiana@cirad.fr


  • • Despite the abundance and diversity of timber tree legumes in the West African rainforest, their ability to form nitrogen-fixing nodules in symbiosis with rhizobia, and their response to rhizobial inoculation, remain poorly documented.
  • • In the first part of this study the occurrence of nodulation was determined in 156 leguminous species growing in six natural forest areas in Guinea, mostly mature trees. In the second part, an in situ experiment of rhizobial inoculation was performed on eight selected tree species belonging to three genera: Albizia, Erythrophleum and Millettia.
  • • Of the 97 plant species and 14 genera that had never been examined before this study, 31 species and four genera were reported to be nodulated. After 4 months of growing in a nursery and a further 11 months after transplantation of plants to the field, we observed a highly significant (P < 0.001) and positive effect of inoculation with Bradyrhizobium sp. strains on the growth of the eight tree species tested.
  • • The importance of determining the nodulation ability of unexplored local trees and subsequently using this information for inoculation in reforestation programmes was demonstrated.


The Leguminoseae is the most represented botanical family in terms of specific diversity among trees of the West African natural rainforest, as recorded in the Côte d’Ivoire where 26% of all commercial timber species are legumes (Dupuy et al., 1997). A large majority (75%) of the timber legume species present in the Guinean rainforest, as observed in our study, are represented throughout all humid zones of West and Central Africa (Lock, 1989). Due to their nitrogen-fixing symbiosis with rhizobia, many native leguminous trees could play an important role in the restoration of N-depleted soils, and might be used as priority pioneer species for the rehabilitation of degraded and overexploited rainforests. However, despite their abundance, diversity and economic importance as high-value timber species in West Africa, very few have been observed for their ability to nodulate and fix atmospheric N symbiotically with rhizobia (Allen & Allen, 1981; de Faria et al., 1989; Wester & Högberg, 1989). Such unexplored symbiotic associations putatively include a high genetic diversity of associated microorganisms, as shown by recent findings describing new species of rhizobia or new genera of legume-nodulating bacteria (Chen et al., 2001; Moulin et al., 2001; Sy et al., 2001). It is considered that only 20% of the known species of Leguminosae worldwide have been examined for nodulation (Sprent, 1995). Nevertheless, studies on nodulation and N-fixing status of tree species are more advanced in other tropical humid regions, especially in the Amazonian basin (Norris, 1969; Bradley et al., 1980; de Faria et al., 1989; Moreira et al., 1992; de Faria & de Lima, 1998; Guehl et al., 1998; Moreira et al., 1998; Roggy & Prévost, 1999). The leguminous woody species found in the latter region are also predominant, representing 40% of the total phytomass (Puig et al., 1990) as well as the most diversified botanical family (Sabatier & Prévost, 1990).

In this context, and because of the lack of information on the N-fixing status of most legume trees native to the West African rainforest, a large and systematic survey of the occurrence of nodulation was carried out on all these species encountered in six different natural forest areas in the south of Guinea. In the second part of our study, the N2-fixing potential of some of the species surveyed, which were chosen for their economic interest, good silvicultural potential and profuse seed production, was evaluated after isolation of rhizobial isolates from the nodules collected in the different sites. Thus, in order to test the effect of rhizobial inoculation on plant growth, a two-step in situ experiment set up in the nursery and in the field was performed on eight species belonging to three different genera: Albizia adianthifolia, Albizia altissima, Albizia ferruginea, Albizia zygia, Erythrophleum guineensis, Erythrophleum ivorensis, Millettia rhodantha and Millettia zechiana.

Materials and Methods

General survey of nodulation status of leguminous trees, shrubs and vines in natural conditions

The survey of nodulation was performed in six different stations of natural forest in the south-eastern part of Guinea: Béro, Diécké, Monts Nimba, Ziama, Pic de Fon and N’zérékoré. All these stations are located in the climatic zone of the humid dense forest from 7°30′ to 9°30′ latitude, with annual rainfall varying from 1800 to 2200 mm. The total area of these protected forests varies from 13 000 to 120 000 ha, and the altitude from 400 to 1300 m.

Before the field survey an exhaustive list of legume species was prepared from those described in a relevant flora (Adam, 1971). The general taxonomy used in this study follows Polhill & Raven (1981), although it was recently modified at the tribe level by Bruneau et al. (2001). The field observations were carried out without any particular sampling protocol, and roots of the targeted species were systematically observed for nodulation when encountered. Herbarium specimens represented by whole plants or different plant samples were collected and are now kept in the herbarium of the Institut de Recherche Agronomique de Guinée in Seredou.

Effect of rhizobial inoculation on the growth of Albizia, Erythrophleum and Millettia spp. in nursery and field conditions

Isolation of bacterial strains  Nodules were collected from mature A. adianthifolia, A. ferruginea, A. zygia, E. guineensis, M. rhodantha and M. zechiana growing in the natural forests of the Ziama reservation (details in Table 1). After 1 month of storage in tubes containing silica gel, the nodules were rehydrated by immersion for 30 min in sterile water, surface sterilized in 30% H2O2 for 10 min and rinsed 10 times in sterile water before being crushed. The crushed nodules were transferred into Petri dishes containing a yeast extract mannitol medium (Vincent, 1970) for bacterial isolation. After incubation at 28°C for 2 wk and several purification subcultures on the same culture medium, seven rhizobial isolates were finally obtained: STM 916 from A. adianthifolia; STM 922 from A. ferruginea; STM 923 from A. zygia; STM 934 from E. guineensis; STM 931 from M. rhodantha; and STM 851 and STM 926 from M. zechiana. These different bacterial isolates formed N-fixing nodules after inoculation to Macroptilium atropurpureum in in vitro monoxenic conditions of culture (Trinick et al., 1991) and were confirmed as Bradyrhizobium sp. strains in previous experiments (Munive, 2002).

Table 1.  Observation of nodulation in leguminous species of the primary rainforest in Guinea (West Africa)
Subfamily, tribe, speciesPlant habit; height (m)LocationNodulationNewly observed species§Newly observed genus
Bussea occidentalis Hutch.T; 25ZiamaYesYes
Chidlowia sanguinea HoyleT; 25Ziama+YesYes
Delonix regia (Hook.) Raf.T; 12NimbaNo*No
Erythrophleum ivorense A. Chev.T; 40Ziama+No**No
Erythrophleum suaveolens (Guill. & Perr.) BrenanT; 40Béro+No**
Mezoneuron benthamianum BaillonS; 5DiéckéYesNo
Cassia aubrevillei Pellegr.T; 15NimbaYesNo
Cassia sieberiana DC.T; 15BéroYes
Chamaecrista kirkii (Oliver) StandleyH; 1.5NimbaNo**No
Chamaecrista mimosoides (L.) GreeneH; 1.5Nimba+No**
Dialium aubrevillei Pellegr.T; 30ZiamaYesNo
Dialium dinklagei HarmsT; 20DiéckéYes
Dialium guineense Willd.T; 15BéroNo*
Dialium pobeguinii Pellegr.T; 15ZiamaYes
Distemonanthus benthamianus BaillonT; 35ZiamaYesYes
Duparquetia orchidacea BaillonS; 8NimbaYesYes
Senna alata (L.) Roxb.S; 5DiéckéNo*No
Senna occidentalis (L.) LinkH; 1.5NimbaNo*
Senna podocarpa (Guill. & Perr.) LockS; 5NimbaYes
Senna siamea (Lam.) Irwin & BarnebyT; 20BéroNo*
Bauhinia thonninguii Schum.T; 9Pic de FonYesNo
Griffonia simplicifolia (DC.) BaillonS; 5N’zérékoreYesYes
Afzelia africana Pers.T; 30BéroNo*No
Afzelia bella var.gracilior KeayT; 35ZiamaYes
Copaifera salikounda HeckelT; 30ZiamaYesNo
Daniellia ogea (Harms) HollandT; 30BéroYesNo
Daniellia thurifera BennettT; 35ZiamaYes
Detarium heudelotianum BaillonT; 20NimbaYesNo
Detarium macrocarpum HarmsT; 25DiéckéNo*
Detarium senegalense J. F. Gmel.T; 25BéroYes
Guibourtia copallifera BennettT; 25ZiamaYesNo
Guibourtia ehie (Chev.) J. LéonardT; 30ZiamaYes
Guibourtia leonensis J. LéonardT; 30DiéckéYes
Tessmannia baikiaeoides Hutch. & Dalz.T; 10NimbaYesYes
Anthonotha crassifolia J. LéonardT; 20BéroYesNo
Anthonotha fragrans (Baker f) Exell & Hillc.T; 30ZiamaNo*
Anthonotha macrophylla P. Beauv.T; 12DiéckéYes
Cryptosepalum tetraphyllum (Hook. f) Benth.T; 25ZiamaYesNo
Gilbertiodendron bilineatum (Hutch. & Dalz.)T; 15ZiamaYesNo
Gilbertiodendron limba (Scott Elliot) J. LéonardT; 15ZiamaYes
Paramacrolobium coeruleum (Taub.) J. LéonardT; 30ZiamaNo**No
Pelligriniodendron diphyllum (Harms) J. LéonardT; 20ZiamaYesYes
Parkia bicolor A. Chev.T; 30NimbaYesNo
Parkia biglobosa (Jacq.) DonT; 15BéroNo**
Pentaclethra macrophylla Benth.T; 25Ziama+No*No
Adenopodia scelerata (A. Ch.) BrenanV; 30NimbaYesYes
Aubrevillea kerstingii (Harms) Pellegr.T; 30ZiamaYesYes
Aubrevillea platycarpa Pellegr.T; 30Ziama+Yes
Calpocalyx aubrevillei Pellegr.T; 25Ziama+YesNo
Calpocalyx brevibracteatus HarmsT; 25ZiamaYes
Dichrostachys cinerea (L.) Wight & ArnS; 8Béro+No**No
Entada africana Guill. & Perr.T; 12Béro+YesNo
Entada gigas (L.) Fawc. & Rend.V; 30Diécké+Yes
Entada mannii (Oliv.) Tisser.S; 15Ziama+Yes
Mimosa invisa Collad.S; 0.5Diécké+No**No
Newtonia aubrevillei (Pellegr.) Keay ssp. aubrevilleiT; 30ZiamaYesNo
Newtonia duparquetiana (Baillon) KeayT; 25NimbaYes
Piptadeniastrum africanum (Hook. f) BrenanT; 40Ziama+No**No
Tetrapleura tetraptera (Schum. & Thonn.) Taub.T; 20Ziama+YesYes
Xylia evansii Hutch.T; 25Diécké+YesNo
Albizia adianthifolia (Schum.) W. F. WrightT; 20Diécké+No**No
Albizia altissima Hook. f.T; 20Ziama+Yes
Albizia dinklagei (Harms) HarmsT; 20Ziama+Yes
Albizia ferruginea (Guill. & Perr.) Benth.T; 20Béro+No**
Albizia glaberrima (Schum. & Thonn.) Benth.T; 15Nimba+No**
Albizia zygia (DC.) J. f. Macb.T; 20Ziama+No**
Bobgunnia fistuloides (Harms) J.H. Kirkbr. & WiersemaT; 15DiéckéYesNo
Amphimas pterocarpoides HarmsT; 25Ziama+YesYes
Angylocalyx oligophyllus (Baker) Baker f.S; 4NimbaYesYes
Baphia nitida Lodd.T; 10Nimba+No**No
Baphia capparidifolia ssp. polygalacea BrummitT; 8Diécké+Yes
Dalbergia afzeliana G. DonV; 35NimbaYesNo
Dalbergia albiflora Hutch. & Dalz. ssp. albifloraV; 15ZiamaYes
Dalbergia bignonae BerhautV; 25Béro+Yes
Dalbergia dalzielii Hutch. & Dalz.V; 6DiéckéYes
Dalbergia hostilis Benth.V; 15Ziama+Yes
Dalbergia oblongifolia G. DonS; V; 10DiéckéYes
Dalbergia rufa G. DonS; V; 15DiéckéYes
Dalbergia saxatilis Hook. f.S; V; 20NimbaYes
Pterocarpus erinaceus PoiretT; 15Béro+No**No
Pterocarpus mildbraedii Harms ssp. mildbraediiT; 20Ziama+Yes
Pterocarpus santalinoides DC.T; 15ZiamaNo**
Abrus canescens BakerH; 2.5Nimba+No**No
Abrus pulchellus ThwaitesH; 2Ziama+No**
Abrus pulchellus ssp. tenuiflorus (Benth.)H; 2Nimba+No**
Dalbergiella welwitschii (Baker) Baker f.S; V; 20NimbaYesNo
Leptoderris brachyptera (Benth.) DunnS; V; 20Ziama+YesNo
Leptoderris fasciculata (Benth.) DunnS; V; 20Nimba+Yes
Lonchocarpus cyanescens (Schum. & Thonn.) BenthS; V; 30NimbaYesNo
Millettia barteri (Benth.) DunnS; V; 20NimbaYesNo
Millettia dinklagei HarmsS; V; 15Nimba+Yes
Millettia griffoniana BaillonT; 15NimbaYes
Millettia lane-poolei DunnT; 7BéroYes
Millettia lucens (Scott Elliot) DunnS; V; 15BéroYes
Millettia rhodantha BaillonT; 12Ziama+Yes
Millettia warneckei HarmsS; V; 7ZiamaYes
Millettia zechiana HarmsS; 10Ziama+Yes
Platysepalum hirsutum (Dunn) HepperS; V; 20NimbaYesYes
Tephrosia flexuosa G. DonS; 1.5NimbaYesNo
Tephrosia nana Schweinf.H; 2NimbaYes
Indigofera atriceps Hook. f. ssp. atricepsH; 3Nimba+No**No
Indigofera dendroides Jacq.H; 1.3BéroNo**
Indigofera heudeloti Benth. ex Baker var. heudelotiiH; 2.5BéroYes
Indigofera macrophylla Schum. & Thonn.S; V; 10ZiamaYes
Indigofera paniculata Pers. ssp. paniculataH; 1.5ZiamaNo**
Indigofera simplicifolia Lam.H; 1.5BéroNo**
Desmodium adscendens (Sw.) DC.H; 75 BéroNo**No
Desmodium incanum (Sw.) DC.H; 1NimbaNo**
Desmodium ramossimum G. DonH; 1NimbaYes
Desmodium salicifolium (Poir.) DC.H; 1.5Ziama+No**
Desmodium velutinum (Willd.) DC.S; 4NimbaNo**
Droogmansia scaettaiana A. Chev. & SillansH; 2Nimba+YesNo
Calopogonium mucunoides Desv.H; V; 4Nimba+No**No
Canavalia ensiformis (L.) DC.H; V; 4NimbaNo**No
Dioclea reflexa Hook. f.S; V; 15NimbaNo**No
Dolichos dinklagei HarmsS; 1.5NimbaYesNo
Dolichos nimbaensis SchnellH; 1.5Nimba+Yes
Dolichos tonkouiensis PortèresS; 1.5Nimba+Yes
Eriosema glomeratum Hook. f.H; 1.5Nimba+No**No
Eriosema parviflorum ssp. collinum HepperH; 0.25Nimba+Yes
Eriosema parviflorum E. Mey. ssp. parviflorumH; 1.5NimbaNo**
Erythrina milbraedii HarmsT; 30Ziama+YesNo
Erythrina senegalensis DC.S; 8Diécké+No**
Glycine wightii Verdc. ssp. wightiiH; V; 5NimbaNo**No
Mucuna flagellipes Hook. f.S; V; 5BéroYesNo
Mucuna poggei Taub.S; V; 20Béro+No**
Mucuna pruriens (L.) DC. var. pruriensV; 5BéroNo**
Mucuna pruriens var. utilis (Wall.ex Wight) Bak. ex BurckV; 10 mNimbaNo**
Mucuna sloanei Fawc. & Rend.H; V; 8NimbaNo**
Physostigma venenosum Balf.S; V; 6BéroYesYes
Rhynchosia brunnea Baker f.H; V; 5Diécké+YesNo
Rhynchosia mannii BakerH; V; 8DiéckéYes
Rhynchosia minima (L.) DC.H, V; 2BéroNo**
Rhynchosia pycnostachya (DC.) MeikleH, V, 8NimbaYes
Teramnus micans (Baker) Baker f.H; V; 4NimbaYesNo
Vigna gracilis (Guill. & Perr.) Hook. f.H; V; 4ZiamaNo**No
Vigna multiflora Hook. f.H; V; 6NimbaYes
Vigna nigritia Hook. f.H; V; 3NimbaYes
Vigna racemosa (G.Don) Hutch. & Dalz.H; V; 6Nimba+No**
Vigna reticulata Hook. f.H; V; 6Nimba+No**
Vigna unguiculata (L.) Walp.H; V; 4Nimba+No**
Vigna venulosa BakerH; V; 2Nimba+Yes
Vigna vexillata (L.) A. Rich.H; V; 4Nimba+No**
Aeschynomene pulchella BakerH; 0.5ZiamaYesNo
Aeschynomene sensitiva Sw.H; 2.5Ziama+No**
Cyclocarpa stellaris BakerH; 0.5NimbaNo**No
Kotschya lutea (Portères) HepperS; 2Nimba+YesNo
Kotschya ochreata (Taub.) Dewit & Duvign. var. ochreataS; 3 mNimba+Yes
Ormocarpum megalophyllum HarmsS; 1. 5NimbaYesNo
Zornia glochidiata DC.H; 0.6Nimba+No**No
Zornia latifolia SmithH; 1Nimba+No**
Crotalaria cylindrocarpa DC.H; 2Ziama+YesNo
Crotalaria doniana BakerH; 2BéroYes
Crotalaria lachnosema StapfH; 1.75Béro+Yes
Crotalaria lathyroides Guill. & Perr.S; 2Béro+No**
Crotalaria spectabilis RothH; 1.25BéroNo**

Plant material  Seeds of A. adianthifolia, A. altissima, A. ferruginea, A. zygia, E. guineensis[= suaveolens (Guill. & Perr.) Brenan], E. ivorensis, M. rhodantha and M. zechiana were collected from adult trees grown naturally in Ziama reservation area. Each of the three genera represented here belong to a different subfamily of the Leguminoseae: Albizia to Mimosoideae; Millettia to Papilionoideae; Erythrophleum to Caesalpinioideae. The germination pretreatment consisted of immersing the seeds in warm tapwater followed by soaking for 6 h.

Inoculation procedure and plant growth conditions in nursery  Inoculation of plants was performed in the experimental nursery of Centre de Recherche Agronomique de Seredou located within the Ziama natural forest area. At the end of the dry season in early January 2001, seeds were directly sown into 12 × 17 cm (diameter × height) polybags filled up with nonsterile forest topsoil from Ziama. The germination of Albizia, Millettia and Erythrophleum species occurred 1, 2 and 3 wk after sowing, respectively. All plants were inoculated 1 month after sowing. The bacterial inoculants consisted in 7-d-old pure rhizobial cultures grown on yeast mannitol medium (Vincent, 1970), and each plant received 1 ml culture containing 109 cells. Except for M. zechiana and A. altissima, which were inoculated with two and three different strains of Bradyrhizobium sp., respectively, each plant species was inoculated with a homologous bacterial strain, or a strain isolated from a species of the same genus when no homologous strain was available. Thirty plants were used per plant species ×Bradyrhizobium strain association, as well as for the control treatments represented by uninoculated plants. Treatments were spaced apart to prevent cross-contamination. After inoculation with rhizobia, plants were grown in the shade for 4 months while shoot height and diameter at collar level were measured on all plants at 1 month intervals.

Experimental design of field trial  The seedlings of the eight species used for the inoculation experiment described above and grown for 4 months in the nursery were transferred directly to the field close to the nursery in Seredou in early June, this date corresponding to the beginning of the long wet season. The open planting area was prepared by hand-cleaning, and no fertilization treatment was applied to the plants. For each tree species, lines of 30 plants were planted per inoculation treatment. Each species tested, except A. altissima and M. zechiana, was represented by one line of 30 trees inoculated with a single Bradyrhizobium strain and one line of 30 uninoculated control trees planted just beside. For A. altissima and M. zechiana, which were inoculated with two and three Bradyrhizobium strains, respectively, one line of 30 uninoculated control trees and two or three adjacent lines of 30 inoculated trees each were planted. Tree spacing was 2 m between trees of the same line, and 4 m between each line. All tree species and inoculation treatments were planted within the same trial, making a total area of 0.46 ha (30 trees per line × 19 lines).

Collection of growth data and statistical analysis  One month after inoculation and every month during the 4 month growing period in the nursery, shoot height and stem diameter of plants were measured on 30 replicates per inoculation treatment (uninoculated control plants and plants inoculated with the different Bradyrhizobium strains) in the eight tree species tested. The height of trees and their stem diameter at ground level were recorded 11 months after field transplantation.

The different inoculation treatments were compared through a one-way anova using the Statistical Analysis System computer program (SAS Institute, 1985). The means obtained from the different inoculation treatments applied to A. altissima and M. zechiana were ranked according to the Newman and Keuls multiple range test (Dagnélie, 1969).

Results and Discussion

Occurrence of nodulation in natural forest conditions

The occurrence of nodulation was investigated in 156 species belonging to different subfamilies and tribes of the Leguminoseae, growing in diverse locations of the humid dense forest in Guinea. The different species observed and their characteristics are listed in Table 1. About 40% of the species analysed were represented by tall mature trees (mean height of observed trees = 22.3 m), while 12% were shrubs, 16% vines, and 32% herbaceous species. Among the Papilionoideae subfamily, 40 out of 90 species (45%) belonging to 10 different tribes were found to be spontaneously nodulated in natural conditions of Guinean forest. Members of this subfamily were represented by the same proportion of ligneous and herbaceous species in our survey (50 and 49%, respectively). No difference in nodulation occurrence was found between both plant habits, with about half the species examined being nodulated in each plant type. The only species observed belonging to the tribe Swartzieae was not nodulated, while the nine other tribes contained nodulated species. A higher proportion of species of the Mimosoideae subfamily, mostly represented by tree species (75%), was nodulated: it occurred in 17 out of 24 species (71%) distributed across the four tribes observed. By contrast, with four positive observations among 42 species analysed (81% of tree species), a low proportion (12%) of species belonging to the Caesalpinioideae was nodulated, and restricted to the Caesalpinieae and Cassieae tribes. Nodulation was not observed in the three other Caesalpinioideae tribes: Cercideae, Detarieae and Amherstieae.

Our observations are consistent with those reported in the literature: we found a high proportion of nodulated species among the Papilionoideae and Mimosoideae and a minority within the Caesalpinioideae. Overall, nodulated species represent 90, 97 and 23%, respectively, of the total number of species observed in these three subfamilies (Allen & Allen, 1981; de Faria et al., 1989; Sprent, 1995). The proportions of nodulated species obtained in our survey were lower than those reported in the literature, especially in the Papilionoideae subfamily, as spontaneous nodulation is often inhibited by different edaphic or climatic factors in natural conditions. So far, nodulation in the Leguminoseae family has been examined in only ≈60% of genera and 20% of species, those remaining unexplored being mostly tropical (Sprent, 1995). Although the trees surveyed in the present study are tall, well known and quite frequent in their natural habitat, a large majority (72%) had never been observed for nodulation before this study. Of the 97 new species and 14 new genera that had never been examined before, 31 species and four new genera are reported here to be nodulated (Table 1). Thus this study shows that an exhaustive list of nodulated leguminous species is far from being finalized, and can easily be expanded by new observations carried out in unexplored biotopes. Within the Caesalpinioideae subfamily, Chidlowia sanguinea, the only species of this genus in the Caesalpinieae tribe, is reported here to nodulate for the first time. According to Sprent (2001), only eight genera within this subfamily have been described and confirmed as nodulated, seven of them belonging to the Caesalpinieae tribe and the other one, Chamaecrista, to the Cassieae. Some nodulation data from other genera of Caesalpinieae were obtained more recently, but need to be confirmed. The nodulation status of the genus Senna, formerly included with Chamaecrista in the genus Cassia, is more controversial and is subject to conflicting reports (Allen & Allen, 1981). The exclusive occurrence of nodulation in the Caesalpinieae and Cassieae tribes among the Caesalpinioideae subfamily was shown to be related to their taxonomic position, which is close to that of the Mimosoideae and Papilionoideae subfamilies. In addition, the absence of nodulation in the other Caesalpinioideae tribes is explained by their inclusion in well differentiated and distant clades, as found by phylogenetic studies based on the analysis of the chloroplast gene rbcL (Doyle et al., 1997). Among the Mimosoideae, Aubrevillea platycarpa and Tetrapleura tetraptera are also the first tree species of new genera that had never been described as being nodulated. The other important tree species of the Mimosoideae subfamily reported to be nodulated for the first time belong to the following genera: Pentaclethra, Calpocalyx, Entada, Xylia and Albizia (see the corresponding species in Table 1). Lastly, among the Papilionoideae, nodulation had never been reported in the genus Amphimas, nor at the specific level in 21 species belonging to 14 other genera.

Effect of rhizobial inoculation on tree growth in nursery and field experiments

As reported in Table 2, the effect of inoculation on plant growth was significantly positive for all species grown in nursery conditions. Positive effects on shoot height and stem diameter were observed as little as 1 month after inoculation and at each time of measurement until the fourth month of nursery growing, just before transferring plants to the field (only shoot heights recorded 4 months after inoculation are shown in Table 2). After 4 months’ nursery growing we observed a significant and positive effect of inoculation with Bradyrhizobium on both stem height and diameter at collar level (P < 0.001 for both parameters) in M. rhodantha (+62.5 and +53.5%, respectively, compared with uninoculated control plants); E. ivorensis (+37.3 and +39.7%); E. guineensis (+20.3 and +27.7%); A. adianthifolia (+69.7 and +94.0%); A. ferruginea (+35.3 and +37.1%); and A. zygia (+58.8 and +29.3%). We also observed a significant and positive effect of inoculation, as well as a variability in the efficiency of the different strains tested, in A. altissima (+19.7% in height and +50.6% in diameter with the most efficient strain, STM922) and M. zechiana (+113.0% in height and +43.8% in diameter with the most efficient strain, STM851).

Table 2.  Effect of rhizobial inoculation on the growth of eight leguminous tree species in the nursery 4 months after inoculation and 11 months after transfer to the field
Tree speciesRhizobium strain treatmentPlant height 4  months after inoculation (cm)Plant height 11  months after field transfer (cm)Stem diameter 11 months after field transfer (mm)
  1. Means ± SE were calculated from 30 replicates per treatment. One-way anova showed significant effects of inoculation on height and diameter at P < 0.01 in each tree species both in the nursery and 11 months after transfer to the field. Means followed by different letters are significantly different according to the Newman and Keuls test at P = 0.05.

Albizia adianthifoliaSTM 91614.05a± 0.2530.1a± 1.5 9.8a± 1.1
Uninoculated 8.28b± 0.2622.6b± 2.3 8.4b± 1.2
Albizia altissimaSTM 91611.82a± 0.6326.7a± 2.2 9.6a± 1.4
STM 92212.04a± 0.3126.6a ± 2.0 9.5a± 1.7
STM 92311.89a± 0.3826.8a± 2.7 8.0b± 1.2
Uninoculated10.06b± 0.3824.4b± 2.3 6.7c± 1.2
Albizia ferrugineaSTM 92220.50a± 0.4559.4a± 8.212.4a± 2.2
Uninoculated15.15b± 0.4230.0b± 3.2 9.7b± 1.5
Albizia zygiaSTM 92314.88a± 0.3950.0a± 3.3 9.0a± 1.4
Uninoculated 9.37b± 0.2326.0b± 3.7 7.8b± 0.7
Erythrophleum guineensisSTM 93422.37a± 0.7657.3a± 8.911.7a± 1.5
Uninoculated18.60b± 0.5330.8b± 2.610.0b± 1.1
Erythrophleum ivorensisSTM 93417.23a± 0.4638.2a± 3.110.2a± 1.4
Uninoculated12.55b± 0.3829.1b± 2.5 8.4b± 0.9
Millettia rhodanthaSTM 93126.10a± 1.0259.5a± 9.311.4a± 1.0
Uninoculated16.00b± 0.6532.7b± 5.210.1b± 1.1
Millettia zechianaSTM 85127.12a± 1.0762.5a ± 11.011.8a ± 1.2
STM 92618.47b± 0.6839.7b ± 8.611.0b ± 1.5
Uninoculated12.73c± 0.4531.6c ± 2.6 9.6c ± 1.1

As indicated in Table 2, 11 months after transplantation of inoculated plants to the field, the positive effects of inoculation observed in the nursery remained significant in all species. The differences in height and diameter measured between inoculated plants and uninoculated control plants were similar to those obtained after 4 months’ nursery growing, and the height differences were enhanced in the following species: A. zygia (+92%); A. ferruginea (+97%); E. guineensis (+87%); M. rhodantha (+82%). Conversely, we noted an important decline of the inoculation effect in A. adianthifolia, although it remained significantly positive (+33% in height, +16% in diameter).

The inoculation experiments presented here were performed using rhizobial strains isolated from the trees observed and described in Table 1. Other studies focusing on the effect of rhizobial inoculation on tree legumes species showed very large differences between the growth of inoculated plants and that of uninoculated control plants at the end of the nursery growing period. This was the case with Acacia mangium, a fast-growing tree species mainly planted for pulp production in South-East Asia, which exhibited increments of ≈40–100% in stem height between inoculated and uninoculated plants according to different nursery conditions (Galiana et al., 1998). Similar increments were observed in other plantation species such as Albizia procera, Albizia lebbeck and Leucaena leucocephala grown in nursery conditions (Aryal et al., 1999). However, in contrast with the present study, such large differences were obtained with selected rhizobial strains obtained after several steps of screening performed among an initial large population of isolated strains. In other cases the inoculation effect on plant growth is nil or very low at the end of the nursery stage, as it was found in several dry-zone tree species including Acacia tortilis, Acacia nilotica, Acacia senegal, Faidherbia albida and Prosopis juliflora, even after a preliminary step of rhizobial strain selection (Brunck et al., 1990). The absence of growth response of a given plant species to inoculation is sometimes attributed to a low level of specificity of their associated rhizobia, as it is often observed in host species nodulating with Bradyrhizobium (Turk & Keyser, 1992). In this case, other local Bradyrhizobium strains are able to take over the inoculated strains and contaminate all the plants in nonsterile nursery conditions. However, many plant species associated with Bradyrhizobium sp. strains known to have a wide host spectrum, such as A. mangium (Galiana et al., 2002), can have a marked growth response to inoculation. We observe similar marked inoculation effects in our experiments performed on Albizia, Erythrophleum and Millettia species where associated strains were genetically identified as Bradyrhizobium sp. strains, as attested by sequencing of the 16S−23S ribosomal DNA intertranscript (Munive, 2002).

Most inoculation studies involving tree legume species have been performed on fast-growing species or species traditionally used in pure plantations or agroecosystems. On the other hand, very few data are available on rhizobial inoculation of long-rotation legume trees such as those found in natural ecosystems. Our results, obtained in in situ conditions, show the need to identify the nodulation status of unexplored local trees before inoculating them with rhizobia in the context of plantation projects or reforestation programmes in degraded humid tropical forests.


We are grateful to P. Deschères and G. Ifono (deceased) (Eaux et Forêts, Guinée) for their help in identifying several tree species, to A. Fontana (IRD, Guinée) for his help during field investigations, to N. Perrier (CIRAD, France) for reading and correcting the manuscript, and to Dr J. H. Kirkbride (USDA, USA) for correcting the manuscript and for helpful discussion.