Interactions between a large community of vertebrate frugivore-granivores (including 7 species of large canopy birds, 19 species of rodents, 7 species of ruminants, and 6 species of monkeys), and 122 fruit species they consume, were studied for a year in a tropical rainforest in Gabon.The results show how morphological characters of fruits are involved in the choice and partitioning of the available fruit spectrum among consumer taxa. Despite an outstanding lack of specificity between fruit and consumer species, consideration of simple morphological traits of fruits reveals broad character syndromes associated with different consumer taxa. Competition between distantly related taxa that feed at the same height is far more important than has been previously supposed. The results also suggest how fruit characters could have evolved under consumer pressure as a result of consumer roles as dispersers or seed predators. Our analyses of dispersal syndromes show that fruit species partitioning occurs more between mammal taxa than between mammals and birds. There is thus a bird-monkey syndrome and a ruminant-rodent-elephant syndrome. The bird-monkey syndrome includes fruit species on which there is no pre-dispersal seed predation. These fruits (berries and drupes) are brightly colored, have a succulent pulp or arillate seeds, and no protective seed cover. The ruminant-rodent-elephant syndrome includes species for which there is pre-dispersal predation. These fruits (all drupes) are large, dull-colored, and have a dry fibrous flesh and well-protected seeds.

For the present study, 15 species of pigeons representing the 5 sub-families usually recognized, and 3 species of sandgrouse were examined. The skeleton and musculature of the bill and hyoid apparatus are described. Morpho-functional analyses show that from a key adaptation to the removal and deglutition (without processing) of attached plant items, pigeons would have followed two pathways, one based on the joint muscular control of the movement of the jaws (Columbinae, Treroninae, Gourinae), the other on the separate muscular control (Didunculinae, Otidiphabinae). Sandgrouse would have diverged from this latter, developing the ability to very selectively remove attached plant items as well as to peck particularly small seeds on the ground. Unexpected differences appeared between sandgrouse species which raise eco-ethological problems.

We address the phylogenetic relationships of ten passerine bird species representing the five presently supposed monarchine (family Monarchidae) genera (Terpsiphone, Hypothymis, Elminia, Trochocercus, Erythrocercus) from Asia and Africa, as well as three monarchs from Australasia, three representatives of the related genera Rhipidura and Dicrurus, and 20 representatives of 11 other oscine groups (including two Culicicapa flycatchers) and one sub-oscine, using two partial mitochondrial genes (cytochrome b and large sub-unit ribosomal 16S RNA). Molecular data corroborate ecological, ethological and morphological observations on the probable heterogeneity of Trochocercus and indicate that this genus is polyphyletic; two of its species are members of Monarchidae allied to Terpsiphone and Hypothymis; the others are more closely related to Elminia. Elminia is not a member of Monarchidae and is not related to any other sampled species, except Culicicapa. Erythrocercus is also outside the Monarchidae but inside a Sylvii-Pycnonotidae group. These results point once more to the need of a fully revised phylogeny of passerine birds. To cite this article: É.

No abstract