In the Sahel, pearl millet yields are affected by the proportion of hybrid phenotype plants resulting from genetic mixing between domesticated and wild forms. Man counteracts this mixing by applying a production method, the efficiency of which is quantified in this study. Under experimental conditions, cultivated and wild pearl millet were hybridised in order to obtain cultivated pearl millet seeds including a known proportion of F1 hybrids tagged by two different allozymes. These seeds were sown in the field and the cultivation was conducted following practices common in the Sahel. The evolution of the survival rate of plants and the frequency of hybrids were followed over several stages during the season: sowing, germinating, emergence, thinning, flowering and maturing of the seeds. Owing to plant mortality in the experiment, the average tendency was a hybrid frequency that decreased steadily in the first part of the growing season from 42% during germination, to 37% at emergence. It then fell to 17% after the thinning of the plantlets by the farmer. At the end of the cycle, after thinning, only 11% of mature plants were hybrids. Thus, under the combined pressures of natural and human selection, the frequency of hybrids in the field declined drastically. In interaction with natural pressure, the farmer's practices of selection of seeds, sowing in pockets and thinning have the combined effect of heavily selecting thecultivated genotype and limiting without completely preventing the introgression of wild pearl millet genes into the cultivated genome.

Molecular cytogenetic analysis has indicated that Coffea arabica is an amphidiploid formed from the hybridization between two closely related diploid progenitor species, C. canephora and C. eugenioides. Our aim was to determine the mode of inheritance in C. arabica and in a tetraploid interspecific hybrid (called arabusta) between C. arabica and C. canephora as revealed by segregation analyses of restriction fragment length polymorphism (RFLP) loci markers. The observed RFLP allele segregations in an F(2) progeny of C. arabica conform to disomic inheritance as expected, with regular bivalent pairing of homologous chromosomes in the F1 hybrid. In contrast, RFLP loci followed tetrasomic inheritance in the arabusta interspecific hybrid, although bivalents have been reported to predominate greatly at meiosis in its hybrid. These results suggest that homologous chromosomes do not pair in C. arabica, not as a consequence of structural differentiation, but because of the functioning of pairing regulating factors. Moreover, the arabusta hybrid seems to offer the possibility of gene exchange between the homologous genomes.

Intergeneric hybrids between Psilanthus ebracteolatus (2n = 22) and Coffea arabica (2n = 44) were successfully produced by crossing at the tetraploid level. Although a total of 41 plants was obtained, only nine plants survived after 5 months growth in the nursery. Hybrid status was confirmed by means of cytological and molecular methods. For most of the morphological characteristics analyzed, hybrids appeared intermediate between the two parental species. The mean production of two surviving hybrids per 100 pollinated flowers as well as their fertility are comparable with those reported for intrageneric crosses between subgenera Coffea species, and do not support the present division of coffee trees into two genera. Both the capacity of C. arabica to hybridize with P. ebracteolatus and the fertility of hybrids produced appear high enough to envisage intergeneric gene transfers from P. ebracteolatus into C. arabica.Key words: coffee, fertility, hybridization, intergeneric, polyploid, wide cross.

Cross-compatibility behaviour of doubled haploid (DH) and hybrid genotypes of Coffea camphora was established using both phenotypic bioassay and in situ seed-set examination. The availability of DHs provided the opportunity of working with genetically homogenous pollen and female parents. The aniline blue fluorescence (ABF) method was applied to detect callose accumulation in pollen and pistil. Clear cross-compatibility/incompatibility situations were observed and confirmed by in situ seed-set analysis. Cross-compatibility analysis of hybrid combinations involving different DHs corroborated the crossing behaviour observed at the DH level. Expression of the self-incompatibility system did not appear to be affected by the low vigour of the DH. The crossing-behaviour distribution observed within DHs derived from clone IF200 confirmed that self-incompatibility in C. canephora is a gametophytic self-incompatibility system controlled by a single locus (S-locus). Reduced seed-set developments following incompatible crosses may indicate the occurrence of pseudo-incompatibility. Molecular marker linkage analysis showed that the S-locus is associated with an RFLP marker on linkage group 9. The availability of a linked DNA marker should facilitate the genetic analysis of self-incompatibility in relation to coffee breeding programmes.

Fifty-five doubled haploids (DH) of Coffea canephora were crossed with either heterozygous genotypes or DH in order to study their combining ability. Three agronomic trials were established. Marked hybrid vigour was observed for all characters analyzed including yield. Large differences were evident among top-crosses involving different DH produced from the same parental clone reflecting the high level of heterozygosity of clones. Factorial mating design analysis indicated that all genetic variance was attributable to additive effects in estimates of yield as well as plant height and leaf characteristics. The general combining ability variance component was also predominant for stem girth and susceptibility to leaf rust, although effects due to interaction were detected. Some hybrid combinations had yield comparable to standard clonal varieties. The implications of such results for breeding of Coffea canephora are discussed. Particularly, the development of F1 hybrid varieties is envisaged.

Doubled haploids (DH) of Coffea canephora Pierre were developed using haploid embryos which occur spontaneously in association with polyembryony. The frequencies of polyembryonic seeds and haploid embryos varied according to the parental genotypes. However, production of a large number of DH seemed possible from all genotypes. More than 750 DHs produced from various genotypes were grown under field conditions and evaluated for different characters of agronomic importance. Approximately half of DH genotypes did not survive, suggesting a strong, negative effect of homozygosity. Inbreeding depression is particularly severe on general vigor and reproductive aspects. For several characters studied such as leaf shape, leaf rust resistance and hundred bean weight, considerable genetic variations were observed within and between groups of DHs constituted by the DHs produced from the same clone. Despite their low vigor and reduced fertility, the DHs of C. canephora offer new possibilities in genetic research and coffee breeding.

Des prospections de cafeiers sauvages, financees par la CEE ont ete effectuees par l'ORSTOM et l'IRCC de 1984 a 1987 en Cote d'Ivoire (region de Danane a l'Ouest, Centre-Ouest et Sud-Est) et en Guinee forestiere (forets de Diecke, Pine, Seredou et Kenema), afin de completer la representation des principales especes en collection au Centre de Ressources Genetiques de Cafeiers Africains en Cote d'Ivoire. Les prospections ont permis de collecter les 4 especes endemiques a la region occidentale de l'aire naturelle de dispersion des cafeiers : C. canephora, C. liberica, C. humilis et C. stenophylla, ainsi que des genres voisins. Des difficultes de prospection de materiels sauvages ont ete ressenties. Elles sont dues a la deforestation qui a reduit l'ampleur ou fait disparaitre des populations de cafeiers et aux hybridations naturelles avec des cafeiers introduits qui ont ete cultives a proximite des reliquats de forets. L'originalite guineenne du materiel a ete verifiee par l'electrophorese d'enzymes. La reconnaissance des genotypes a permis ainsi pour C. canephora d'identifier 86 guineens purs issus de 6 nouvelles populations