**7. Association with traditional farming systems**

which proved to be much less diverse than modern barley cultivars due possibly to their relative geographic isolation, their relatively recent introduction to Tibet and the fact that they have been subject to very little natural or man-made selection [43]. Therefore, the dynamics of genetic diversity and changes over time of the genetic structure of landraces are likely to be crop specific. It is also likely to be associated with the mode of fertilization (self- versus cross) and propagation (sexual or asexual), which has over time resulted in genetic bottlenecks, varying outcrossing rates, recombination and gene flow. Thus, as Almekinders and Louwaars [24] conclude, 'a landrace is usually a complex heterogeneous population, but not necessarily so'.

Landraces are generally adapted to local environment. With the continued cycles of local planting, harvesting and farmers' selection, over time landraces will be selected for local environmental and agroecosystem conditions and practices, just as ecotypes of wild species are adapted to the local environmental conditions. Landraces 'are adapted to their growing conditions' [11]; 'possess adaptive complexes associated with the special conditions of cultivation, pure-stand associations, harvesting and others factors' [44]; 'are not only adapted to their environment, both natural and man-made, but they are also adapted to each other' [7]; 'are adapted to the areas in which they grow' [12]; 'are specifically adapted to local conditions'

Bennett [44] made the assumption that landraces are more suited to cultivation in particular locations than highly bred cultivars that are bred for cultivation in the most common environmental conditions. Inevitably, cultivars will be less suited to grow in suboptimal conditions and therefore have less of a competitive advantage in marginal environments where the local landraces are likely to have an adaptive advantage. These local conditions may be defined as abiotic (e.g. salinity, drought, etc.), biotic (e.g. pests, diseases, weeds) and human (e.g. cultivation, management and use). Landraces are perceived to have the ability 'to sensitively respond to even minor environmental influences' [44]; 'to have some built-in insurance against hazards' possibly due to their inherent population structure [7]; 'to accumulate resistance genes to limiting factors in the physical and biological environment—drought, cold, diseases, pests' [24]; and 'to be capable of producing in any but disaster seasons at a level which safeguards

Several studies have demonstrated the relationship between landraces and local adaptation, for example; Frankel [8] and Brown [39] discuss landrace adaptation to marginal conditions associated with climatic, soil and disease stress. The evolution of local adaptation over millennia in these stressed environments ensures yield stability even in extremely adverse years. In this sense, Zeven [27] considers yield stability to be a principal characteristic of

However, even though there are numerous references to a specific relation between a landrace and local environmental conditions, there are exceptions. Zeven [3] indicated that 'some landraces are able to adapt themselves to a wide range of environments, whereas others are able to

**6. Local genetic adaptation**

10 Rediscovery of Landraces as a Resource for the Future

[13]; and 'are adapted to local conditions' [26].

landraces.

the survival of the cultivator' and so provide yield stability [24].

Traditional farming systems have often been considered beneficial reservoirs of landraces and intra-crop diversity [45]. Traditional farming systems involve traditional cultivation, storage and use practices, and integrated with these practical skills, traditional knowledge about landrace identification, cultivation, storage and uses is incorporated. In this sense, one important element of landraces conservation that has recently been the focus of researchers' attention is the way that landraces studies have focused on farmers' variety selection [46], farmers' seed exchange [22], farmers' seed networks [22], farmers' seed replacement [20], farmers' portfolios of varieties [19], farmers' landraces identification [47]and farmers' landrace uses [48]. Each has shown the role of farmers for the creation and maintenance of a landrace.

In fact, Zeven [27] suggested that landrace diversity can be explained by the combination of farmers' selection criteria on specific local landrace genotypes by means of farmers' seed saving and the introduction of variation by means of exchange with other farmers of other genotypes of the same crop. This indicates that landraces are more inherently dynamic than cultivars as they are maintained through repeated cycles of sowing, harvesting and replacing seed selection by farmers [49, 50] within complex informal systems. However, it is also important to consider that traditional farming systems are themselves also dynamic and that the frontier between them and other farming systems is not well defined. As such, traditional farming systems are subject to change, incorporating in some cases modern cultivars into their systems, growing them alongside landraces of the same species [51]. These have been managed and maintained by farmers.
