We examined how forest fragmentation affected the population genetics of the threatened tree species Manilkara maxima from the Atlantic coastal region of southern Bahia, Brazil. Manilkara maxima is economically important as a source of timber and identified as one of the top five “extremely valuable” tree species for conservation of the endangered golden-headed lion tamarin (GHLT) (Leontopithecus chrysomelas). The Brazilian Atlantic forest is considered a global “hotspot” with a high conservation priority due to its abundant biodiversity and forest fragmentation. Forest fragmentation threatens biodiversity by creating smaller spatially isolated populations with a greater probability of extinction. Fragmentation is expected to reduce tree density, allelic diversity, and gene diversity, while increasing inbreeding, and fixation. We hypothesized relationships between forest fragment size and tree density and indices of genetic diversity: allelic diversity, gene diversity, inbreeding, and fixation. We sampled a total of 222 individuals from two large fragments (400 ha and 200 ha) and one site consisting of three 25 ha, two 10 ha, and three 5 ha fragments within a matrix of plantations, fields and regenerating forests. Genomic DNA was extracted from leaves and cambium and five microsatellite loci were used for examining the population genetic diversity.
Results/Conclusions
Our findings suggest larger forest fragments maintain higher tree densities, but medium sized fragments can maintain high levels of genetic diversity for this threatened tree species. As forest size decreased, adult and juvenile tree density decreased, which is alarming considering the long-term sustainability of this tree, since only 5% of forests in this region are > 100 ha. Additionally, large tree density was significantly lower in smaller fragments, suggesting smaller fragments may not be suitable for GHLT, since they prefer larger trees for foraging and sleeping. Juvenile trees exhibited the greatest mean allelic diversity and gene diversity in the 25 ha forest when compared to the largest forest fragments. However, when the 5, 10, and 25 ha fragments were analyzed we found allelic diversity and gene diversity decreased with decreasing fragment size. These results suggest that the 25 ha fragments have higher allelic diversity and gene diversity than the largest forest fragments. This may result from either long distance dispersal from other forests, or dispersal from adult trees within the fragments outside of the sampled area. Inbreeding and fixation were not correlated with fragment size for the juvenile trees.