Résumé
Clonally propagated crop species are less adaptable to environmental changes than those propagating sexually. DNA studies have shown that in all countries where taro (Colocasia esculenta (L.) Schott) has been introduced clonally its genetic base is narrow. As genetic variation is the most important source of adaptive potential, it appears interesting to attempt to increase genetic and phenotypic diversity to strengthen smallholders' capacity to adapt to climatic changes. A global experiment, involving 14 countries from America, Africa, Asia and the Pacific was conducted to test this approach. Every country received a set of 50 indexed genotypes in vitro assembling significant genetic diversity. After on-station agronomic evaluation trials, the best genotypes were distributed to farmers for participatory on-farm evaluation. Results indicated that hybrids tolerant to taro leaf blight (TLB, Phytophthora colocasiae Raciborski), developed by Hawaii, Papua New Guinea and Samoa breeding programmes outperformed local cultivars in most locations. However, several elite cultivars from SE Asia, also tolerant to TLB, outperformed improved hybrids in four countries and in one country none of the introductions performed better than the local cultivars. Introduced genotypes were successfully crossed (controlled crossing) with local cultivars and new hybrids were produced. For the first time in the history of Aroids research, seeds were exchanged internationally injecting tremendous allelic diversity in different countries. If climatic changes are going to cause the problems envisaged, then breeding crops with wide genetic diversity appears to be an appropriate approach to overcome the disasters that will otherwise ensue.