LONG-TERM DYNAMICS OF MEDITERRANEAN FRUGIVOROUSBIRDS AND FLESHY FRUITS: A 12-YEAR STUDY

The relationship between fleshy-fruited plants and their vertebrate seed dispersal agents often has been depicted as subject to important interannual variation, but no study has thus far documented such variation on a long-term basis. This paper presents the results of a 12-yr investigation on fleshy-fruited plants and avian frugivores in a Mediterranean montane locality of southeastern Spain. The main objective was to document patterns and correlates of long-term vari- ation in the composition and abundance of fruits and birds, with particular reference to seed dis- persers. During October-December (''autumn'' period) 1978-1990, abundance of ripe fruits and birds was assessed in a 4-ha plot in dense, well-preserved sclerophyllous scrub, by means of counts in permanent plots and mist-netting, respectively. Diet composition and fruit preference patterns of Erithacus rubecula and Sylvia atricapilla, the two most abundant seed dispersers, were also inves- tigated over the same period, using fecal sample analyses. Possible consequences to the birds of annual variation in fruit supply and diet composition were investigated using data on fat deposition levels and recapture rates of mist-netted individuals. Total fruit abundance (i.e., mean ripe fruit density of all species combined) fluctuated among years between 5.4 6 11.1 fruits/m 2 (mean 6 1 SD; 1986) and 77.1 6 78.0 fruits/m 2 (1989) and was positively related to the amount of rainfall in the preceding spring. Not all fruiting species bore ripe fruits every year, and among those species that did, fruit density fluctuated asynchronously and to variable degrees. Seven out of 13 species exhibited significant supra-annual periodicity in fruit abundance, with fluctuation periods ranging from 2 to 6 yr. Variation in the abundance of each of the six most abundant fruit species was unrelated to annual variation in rainfall. The autumn bird assemblage at the study site was made up of year-round resident species (54.1% of captures, all years combined) that were largely fruit predators (feeding on pulp or seeds without performing dispersal) and of autumn-winter resident species (45.5% of captures) that were largely seed dis- persers. Bird abundance, all species combined, ranged between 27.1 and 61.5 captures/100 net-hours for 1987 and 1986, respectively. Depending on year, seed dispersers made up 25.6-75.4% of the total captures, and fruit predators made up 20.9-69.7%. The relative importance of nonfrugivores was always negligible (1.1-9.9%). No correlation existed across years between total fruit abundance and the capture rates of all bird species combined, seed dispersers, or fruit predators. Annual variation in the abundance of seed dispersers was positively related to November mean maximum temperature. At the individual species level, S. atricapilla capture rates were correlated with the abundance of the fruits of Phillyrea latifolia, a species exhibiting extreme annual fluctuations. Annual variation in the importance of fruits in the diet of S. atricapilla and E. rubecula was not significantly related to changes in fruit abundance. Composition of the fruit diet of these species fluctuated markedly among years, and there was little agreement between composition of the diet and of the fruit supply. Certain fruit species were significantly preferred, and others avoided, by both S. atricapilla and E. rubecula. Ranking of interspecific fruit preferences remained consistent among years and was related to differences in carbohydrate and lipid content of fruit pulp. Fat deposition levels of S. atricapilla and E. rubecula did not vary among years and were not significantly related to fruit abundance, percentage of fruit volume in the diet, or contribution of lipid-rich fruits to the diet. Return rates of individuals of these species to the study locality in successive wintering periods were very low, and not significantly related to diet composition or fruit abundance. Prevalence of abiotic over biotic determinants of annual variations, extensive decoupling of the long-term temporal dynamics of fruits and dispersers, and the remarkable ''indifference'' of frugivores to variations in the fruit supply all point to the non-equilibrial nature of this assemblage of fleshy- fruited plants and their avian dispersers.