References w w w . a g r a r . u n i – k a s s e l . d e 1 Organic Plant Production and Agroecosystems Research in the Tropics and Subtropics (OPATS), University of Kassel, Germany 2 Forest Genetics and Forest Tree Breeding, Georg-August-University Göttingen, Germany 3 Tree Genetic Resources and Domestication, World Agroforestry Centre (ICRAF), Nairobi, Kenya 4 Department of Horticulture, University of Khartoum, Sudan 5 Sustainable Agricultural Production Systems with Special Focus on Horticulture, Rhine-Waal University of Applied Sciences, Germany Martin Wiehle 1 , Kathleen Prinz 2 , Katja Kehlenbeck 3 , Sven Goenster 1 , Seifeldin Ali Mohamed 4 , Reiner Finkeldey 2 , Andreas Buerkert 1 and Jens Gebauer 5 Baobab (Adansonia digitata L.) – Morphological and genetic diversity of a neglected population in the Nuba Mountains, Sudan Diversity: Tree height and diameter at breast height differed significantly among locations, whereas fruit morphology exhibited low differences. However, tree-to-tree variation within locations was high. Samples from the southwest showed slightly higher genetic diversity indices (Table). High diversity in all locations. Genetic structure: A Bayesian clustering approach structured the samples into two genetically different groups (Fig. 3): cluster 1, mainly trees from the southwest (SW) and cluster 2, mainly trees from other locations. Urgent need to implement conservation strategies in both genetically distinct units. Human impact: Baobabs sampled in or close to homesteads showed slightly higher genetic diversity compared to trees far from homesteads (‘wild’). First indication that human-mediated seed admixture from various origins have possibly enriched the genetic base in the homestead stands. Introduction & Methods Results & Discussion Adansonia digitata L. is an important fruit tree of semi-arid sub-Saharan Africa (Fig. 1). Baobab’s fruit pulp provides local nutrition and supports national economies and even export markets. Despite its importance, information of the tree’s ecology and distribution as well as morphological and genetic diversity is lacking, particularly for the East African range. Within that region, Sudan harbors the northernmost populations with potential adaptations to dry conditions (Fig. 1). However, these populations are threatened by lack of rejuvenation as well as habitat loss due to climate change and human impact. A strong human association to baobabs leads to the question if humans altered the species’ variation. Therefore, morphological (dendrometric and fruit traits) and genetic data (microsatellite markers) of 306 trees from Sudan were assessed that can be used to develop suitable ex and/or in situ conservation strategies for this important tree species. Figure 1. Natural distribution map (Wickens, 1979) and sampling locations in the Nuba Mountains, Sudan. Study area The surprisingly high morphological variability and genetic diversity of baobab highlights the potential for domestication of this neglected wild tree in Sudan. The development of region-specific and sustainable management strategies as part of circa situm conservation approaches will contribute to maintain the genetic resources of this important, but threatened species in the Nuba Mountains. [email protected] Figure 3. Bayesian cluster plots obtained with the STRUCTURE software package 2.3.1 (Pritchard et al. 2000) for two clusters, the most plausible grouping of all investigated genotypes. Each vertical bar partitioned into two colored segments refers to a particular tree and illustrates the assignment likelihood of the respective genotype to one of the two clusters. Table. Morphological and genetic indices assessed for 306 A. digitata trees in the Nuba Mountains, 2010. ns=non-significant according to Kruskal-Wallis for locations and Mann-Whitney test for ‘Homestead’ and ‘Wild’. Homestead trees Microsatellite application DNA extraction Wickens, G. E. (1979), Chapter 15: The uses of the baobab (Adansonia digitata L.) in Africa. In: Taxonomic aspects of African economic botany, Kunkel, G. (ed.). Pritchard, J. K., M. Stephens, et al. (2000). Inference of population structure using multilocus genotype data. Genetics 155: 945–959. Financially supported by Figure 2. Baobab leaf and section of DNA double helix. Presented at GTÖ 2013, Wien Kadugli Conclusions